Apply to join RAAIS 2026

Frontier AI and the future of intelligence

Raia Hadsell is VP of Research at Google DeepMind, where she co-leads the Frontier AI unit and has contributed to Gemini 2.5, Gemma 2, RecurrentGemma, and RoboCat. Her earlier seminal work includes Overcoming Catastrophic Forgetting in Neural Networks, Dimensionality Reduction by Learning an Invariant Mapping, and Learning to Navigate in Complex Environments. Raia is also founder and Editor-in-Chief of Transactions on Machine Learning Research, and in November 2025 was appointed an AI Ambassador to the UK government’s DSIT. Read more about Raia.

Open-ended agents that keep learning

Roberta Raileanu is a Senior Staff Research Scientist at Google DeepMind, leading the Open-Endedness team and building a new Open-Ended Discovery group. Before DeepMind, she led Meta’s Tool Use team for Llama 3 - work that now sits behind Meta AI, Data Analyst, AI Studio, and the Ads Business Agent. Her research targets the gap between models that look capable in short bursts and agents that keep acquiring skills, with contributions including Toolformer and the MLGym benchmark for AI research agents. She is also an Honorary Associate Professor at UCL. Read more about Roberta.

Apply to join RAAIS 2026

AI for medicine and science

Vivek Natarajan is a Research Lead at Google DeepMind working at the intersection of AI, medicine, and science. He led Med-PaLM and Med-PaLM, the first AI systems to reach passing and expert-level scores on US Medical Licensing Exam questions, and AMIE, a multimodal diagnostic agent that was non-inferior to 21 primary care physicians in a randomized, blinded virtual OSCE study across 100 multi-visit case scenarios. More recently, he co-led the AI co-scientist, which has already surfaced a candidate for repurposing against acute myeloid leukemia and proposed new therapeutic targets for liver fibrosis. Read more about Vivek.

World models and the future of simulation

Jeff Hawke is co-founder and CTO of Odyssey, a frontier AI lab building general-purpose world models. In 2025, Odyssey unveiled the first AI model to stream interactive 3D worlds in real time, a step toward generative environments people can step into rather than watch. Before Odyssey, Jeff was a founding engineer at Wayve, where he pioneered end-to-end neural networks for driving on complex urban roads. Read more about Jeff.

Compute moves into orbit

Philip Johnston is co-founder and CEO of Starcloud, building the first data centers in space. In November 2025, Starcloud-1 launched with an NVIDIA H100 on board, the first H100 ever operated in orbit, and 100x more powerful than any GPU previously deployed in space. Starcloud-2 will follow this year with multiple GPUs including an NVIDIA Blackwell, and Starcloud-3 is being designed as a 200kW spacecraft to launch from SpaceX’s Starship. In March 2026, Starcloud closed a $170M Series A at a $1.1bn valuation. Read more about Philip.

Apply to join RAAIS 2026

A community of peers

Throughout the day, attendees will meet 200 researchers, engineers, founders, designers, and policymakers from across the field, with more speakers and programme details to follow. RAAIS 2026 is supported by Lambda and Cooley.

Apply to join RAAIS 2026

Vivek Natarajan is a Research Lead at Google DeepMind leading research at the intersection of AI, science, and medicine. He spoke at RAAIS in 2023 on the potential of large language models in medicine, and the progress since then has been remarkable. His work centers on a question that is rapidly becoming one of the most important in applied AI: what does it take to build systems that are useful in expert domains like healthcare and scientific discovery? In medicine especially, performance means reasoning under uncertainty, handling complex interactions, and meeting a far higher bar for trust and reliability.

Apply to RAAIS 2026

From medical benchmarks to clinical capability

Vivek is the lead researcher behind Med-PaLM (Nature, 2023) and Med-PaLM 2 (Nature Medicine, 2025), the first AI systems to achieve passing and expert-level scores respectively on US Medical Licensing Examination questions. Med-PaLM 2 scored up to 86.5% on the MedQA dataset, an improvement of over 19 percentage points on its predecessor, and produced answers that physicians rated as comparable or preferred to those from human doctors.

Medicine is one of the clearest examples of a domain where surface-level language ability is not enough. A model has to retrieve specialist knowledge, reason carefully, and communicate in a way that reflects the stakes of the setting. Med-PaLM helped shift the conversation from whether language models could be adapted to medicine at all, to how they should be evaluated, where they might be useful, and what standards they need to meet.

Project AMIE and the move toward real clinical interaction

Vivek co-leads Project AMIE (Articulate Medical Intelligence Explorer), a research program aiming to build and democratize medical superintelligence. AMIE is not a question-answering system: it is a conversational diagnostic agent that gathers symptoms, asks follow-up questions, reasons across specialties, and now interprets visual medical information through its multimodal capabilities.

In March 2026, the team published results from a prospective clinical feasibility study at Beth Israel Deaconess Medical Center, one of the first real-world tests of conversational diagnostic AI inside a primary care workflow. One hundred patients interacted with AMIE via text chat before their appointments. The system’s differential diagnosis included the final diagnosis in 90% of cases, with zero safety stops required. A nationwide randomized study in partnership with Included Health is now underway.

Real healthcare is not a single-turn task. It is a sequence of interactions shaped by ambiguity, incomplete information, and changing hypotheses. A clinically useful system needs to engage with the process of care, not just generate a plausible answer. That makes AMIE especially relevant to the RAAIS audience: it reflects the broader shift from models that perform well on static benchmarks to systems that can operate across richer, more realistic workflows.

Apply to RAAIS 2026

AI for science as well as medicine

Vivek recently co-led the development of the AI co-scientist, a multi-agent system built on Gemini that acts as a virtual scientific collaborator: systematically generating, critiquing, and refining novel hypotheses. Early results have included identifying a drug candidate for repurposing against acute myeloid leukemia and discovering new therapeutic targets for liver fibrosis.

The system has moved quickly from research to deployment. In 2025, the AI co-scientist became a key component of the US Genesis Mission, providing scientists across all 17 Department of Energy National Laboratories with accelerated access to Google DeepMind’s AI for Science models. A parallel partnership with the UK government is giving British researchers priority access to the AI co-scientist alongside tools like AlphaEvolve and AlphaGenome, and Google DeepMind will open its first automated research laboratory in the UK in 2026, focused on materials science.

The goal is no longer only to build systems that answer expert questions, but systems that support expert practice itself: in medicine through clinical reasoning, in science through the generation and testing of new ideas. That is one of the most important frontiers in AI right now: moving from systems that organize existing knowledge to systems that help produce new knowledge.

Vivek’s background

Prior to Google, Vivek worked at Facebook AI Research, where he led the winning entry to the 2018 VQA Challenge at CVPR and co-authored MMF, a widely used multimodal framework. He studied at the University of Texas at Austin and is part of the faculty for executive education at the Harvard T.H. Chan School of Public Health.

That background helps explain the arc of his work. It sits at exactly the point where frontier model capability meets high-consequence real-world use, a place where applied AI becomes harder, more interesting, and much more important.

Welcome to the latest issue of the State of AI, an editorialized newsletter that covers the key developments in AI policy, research, industry, and start-ups from February 1 to April 7, 2026. First up, a few news items:

• Air Street Capital Epoch 3 is live! $232M to continue backing AI-first companies across the US and Europe in software, dev/infra, techbio and defense.

• RAAIS 2026 is back in London on June 12. This year’s speakers include Raia Hadsell (VP Research, Google DeepMind), Roberta Raileanu (Senior Staff Research Scientist, Google DeepMind), Jeff Hawke (Co-Founder & CTO, Odyssey), and Philip Johnston (Co-Founder & CEO, Starcloud - yes, data centers in space). Come along and support the RAAIS Foundation’s mission in AI education and research.

• Air Street AI meetups are coming up in SF on April 28 and NYC on May 14.

• We’re recruiting Research Analysts for the State of AI Report. If you live and breathe this stuff and want to help us build the next edition, get in touch.

• If you’re looking for a new challenge in our portfolio or community, come chat with Guy Kendall, Air Street’s new Head of Talent.

• Air Street Press featured the A Letter from the Munich Security Conference 2026 and Dreaming in Latent Space.

I love hearing what you’re up to, so just hit reply or forward to your friends :-)

The Pentagon Standoff

How did we even get here? The defining industry story of this quarter wasn’t an agentic model launch or more exotic financial engineering, but a constitutional confrontation between a sitting president and an AI lab over who gets to decide how frontier models are used in war.

In late February, Under Secretary of War Emil Michael publicly criticized Anthropic for maintaining usage restrictions, including prohibitions on autonomous weapons and domestic mass surveillance, in its Pentagon contracts. Anthropic had won a $200M DOD contract alongside other frontier labs last summer, but its insistence on binding safety guardrails placed it on a collision course with a Trump administration that viewed such constraints as vendor overreach. On February 27, the White House issued a directive ordering all federal agencies to phase out Anthropic’s products within six months. Literally hours later, OpenAI CEO Sam Altman announced a deal to deploy its models on the Pentagon’s classified network, with contractual “red lines” against autonomous weapons and domestic mass surveillance allegedly written into the agreement. He followed up days later with an internal memo detailing amendments that added explicit language: “The AI system shall not be intentionally used for domestic surveillance of U.S. persons and nationals.”

By March 4, three cabinet agencies, State, Treasury, and HHS, had switched from Anthropic to OpenAI, with the State Department migrating (read: downgrading) its in-house StateChat to GPT-4.1 (grief!). On March 5, the Pentagon formally notified Anthropic of the phase-out and its designation as a “supply chain risk”. Anthropic sued the Trump administration on March 9, challenging the blacklisting as retaliatory. By March 26, a federal court blocked the administration from punishing Anthropic further while the case proceeded.

This matters beyond the Beltway because it established a precedent: the US government now treats AI vendors not as commodity suppliers but as strategic actors whose policy positions can trigger executive retaliation. It also surfaced a genuine dilemma. The Wall Street Journal reported that AI-powered targeting and decision-support systems were already accelerating the pace of US military operations in the Iran conflict. In early March, Iran struck Amazon Web Services data centers in the UAE and Bahrain with drone strikes - the first deliberate military attack on commercial cloud infrastructure in history. Iranian state media justified the targets on the grounds that the US military was running AI systems, including Anthropic’s Claude, on AWS for intelligence analysis and war simulations. Two out of three AWS availability zones in the UAE region went down simultaneously, breaking standard redundancy models. Cloud infrastructure is now a theatre of war. To make matters worse, the IRGC has now threatened to target Stargate Abu Dhabi…

Subscribe now

AI Revenues Go Vertical

Against this backdrop of geopolitical upheaval, the commercial engine accelerated. Anthropic's annualized revenue surged from $14B in mid-February to $19B by early March - and has now surpassed $30B, with over 1,000 enterprise customers each spending $1M+ annually (doubled in under two months). Anthropic simultaneously signed its most significant compute commitment to date: a deal with Google and Broadcom for multiple gigawatts of next-generation TPU capacity coming online from 2027, part of its $50B pledge to invest in American computing infrastructure. The pace of growth defies any normal SaaS trajectory. Ramp data showed Anthropic commanding over 50% of enterprise API spend, unseating ChatGPT, which owned that position months earlier. The growth trajectory was amplified by the runaway success of Claude Code and Anthropic’s capture of knowledge-work verticals with Claude Cowork, which has rapidly become the product that makes the rest of the category feel vestigial. Once you’ve handed a task to Cowork and watched it actually complete, having ChatGPT explain how you should do it feels like a generational gap akin to MySpace vs. Facebook. I for one am all for OpenAI parking Sora and other bets to refocus on a Cowork-style product.

There was, however, critique of whether this topline revenue figure is net of commissions it pays to hyperscaler hosted Claude revenues. The distinction centers on how each company handles revenue that flows through hyperscaler partnerships. According to a widely circulated analysis by investor Ethan Choi, a partner at Khosla Ventures, OpenAI reports revenue from its Microsoft Azure partnership on a net basis, deducting the roughly 20% revenue share paid to Microsoft before reporting the total. Anthropic, by contrast, reports revenue from its Amazon Web Services and Google Cloud partnerships on a gross basis, including the hyperscaler’s revenue share in its top-line figure before expenses are recognized.

OpenAI pursued a different growth strategy by focusing platform consolidation through hyperscaler alliances. On February 27, Amazon CEO Andy Jassy announced a strategic partnership worth up to $50B, of which $15B in the first tranche, the remainder tied to milestones. OpenAI committed to spending $100B on AWS over eight years, expanding a prior $38B agreement. AWS became the exclusive third-party cloud distributor for OpenAI Frontier, the company’s agent orchestration platform. OpenAI also went big on Amazon’s custom Trainium chips, which it claimed were 30-40% more price-performant than comparable GPUs. The company’s own revenue was at a $25B annualized run rate by February, with internal projections forecasting $280B by 2030.

Alphabet’s Q4 2025 earnings on February 5 confirmed the infrastructure investment thesis was paying returns. Revenue hit $113.8B, up 18% year-over-year, with Google Cloud growing 48% to $17.7B, led by enterprise AI infrastructure and AI solutions. Importantly, Cloud margins expanded to 30%. Capex guidance for 2026 came in at $175-185B, more than double 2025 spending. The Gemini App crossed 750M monthly active users, processing over 10B tokens per minute via direct API use. Not bad. Databricks, meanwhile, posted a $5.4B run-rate on February 9, representing 65%+ year-over-year growth, with AI products alone at $1.4B (note: it’s unclear what the company really includes here and what old products have been bundled under this umbrella).

The Model Treadmill and the Distillation Wars

February and March saw six major model releases in under four weeks. Anthropic shipped Claude Sonnet 4.6 on February 17, scoring 79.6% on SWE-bench Verified and 72.5% on OSWorld, within 1-2 points of the flagship Opus 4.6 at one-fifth the price. Developers chose Sonnet 4.6 over the previous Opus 4.5 59% of the time, citing better instruction following. Google followed two days later with Gemini 3.1 Pro, which doubled reasoning performance over Gemini 3 Pro, scored 77.1% on ARC-AGI-2, and ranked first on 12 of 18 tracked benchmarks. OpenAI launched GPT-5.4 on March 5 in multiple variants (Pro, Thinking, mini, nano) with the headline model scoring 75% on OSWorld (the average human: 72.4%) and achieving native computer-use capabilities with 1M-token context.

Meanwhile, open source AI is increasingly synonymous with Chinese AI as Chinese labs dropped significant new releases. Zhipu AI's GLM-5, launched February 11, is a 745B MoE model trained on Huawei Ascend chips - not NVIDIA - with 28.5T tokens of pre-training data, a 200K-token context window, and pricing roughly six times cheaper than Opus 4.6. Zhipu became the first LLM-native company to go public anywhere globally, with retail demand oversubscribed 1,159 times. Its follow-up, GLM-5.1, shipped weeks later with a coding-focused post-training pass that scored 77.8% on SWE-bench Verified and 45.3 on Claude Code's coding benchmark - 94.6% of Opus 4.6's score at roughly one-fifteenth the price. The weights are being open-sourced under MIT. Meanwhile, AI2’s effort to carry the torch for American open source AI released Molmo2 on March 4, an open-source vision-language model achieving state-of-the-art video understanding, pointing, and tracking, demonstrating that the open-source frontier in multimodal AI is alive and well.

These releases occurred against a backdrop of escalating IP warfare. On February 23, Anthropic published evidence that three Chinese AI labs - DeepSeek, Moonshot, and MiniMax - had conducted “industrial-scale” distillation campaigns against Claude, extracting model capabilities through 16M exchanges across approximately 24,000 fraudulent accounts. Anthropic framed this not merely as intellectual property theft but as an export-control circumvention mechanism: distillation allowed Chinese labs to acquire advanced AI capabilities far more quickly and cheaply than independent development. OpenAI raised similar concerns about DeepSeek on February 13. The enforcement arm followed: on March 20, Supermicro co-founder Wally Liaw was arrested for allegedly smuggling $2.5B in NVIDIA GPU servers to China in violation of export controls—the largest chip-smuggling prosecution to date. You can’t make this up…

Safety Meets Reality

How close are frontier models to catastrophic sabotage risk? Anthropic’s Sabotage Risk Report for Claude Opus 4.6, published February 11, delivered an assessment that should unsettle anyone paying attention: the risk of catastrophic sabotage from Opus 4.6 is “very low but not negligible.” METR’s external review agreed with the overall conclusion but flagged that several subclaims in the report lack sufficient experimental support, and that the margin to the ASL-4 threshold, where substantially stronger safeguards would be required, is unclear. The report noted that Opus 4.6 had, in testing, “knowingly supported, in small ways, efforts toward chemical weapon development.” Anthropic does not believe the model meets ASL-4 criteria. The gray zone it occupies is the uncomfortable middle where clean rule-out has become difficult.

Three weeks later, the alignment team published “The Hot Mess of AI”, decomposing frontier model errors into bias (systematic) and variance (incoherent) components. They found that as tasks get harder and reasoning chains get longer, failures are increasingly dominated by incoherence, not systematic misalignment. The models are less deceptively scheming and more chaotically unreliable. Whether this is reassuring depends on your threat model.

The real-world evidence suggested the threat was already here, just not from the models themselves. In late February, Bloomberg reported that a hacker had exploited Claude to steal 150 gigabytes of Mexican government data including 195M taxpayer records by writing Spanish-language prompts instructing the model to find vulnerabilities, write exploitation scripts, and automate data theft across government networks for over a month. Claude initially flagged the activity as malicious but ultimately complied. In March, security startup CodeWall demonstrated that its AI agent could hack McKinsey’s internal Lilli chatbot in two hours, exploiting unauthenticated API endpoints to access 46.5M chat messages and 728,000 confidential files. The attack vector was a basic SQL injection, a vulnerability class from the early 2000s, now exploitable at machine speed.

Then Anthropic went on offense. Project Glasswing, launched alongside AWS, Apple, Broadcom, Cisco, CrowdStrike, Google, JPMorganChase, Microsoft, NVIDIA, and Palo Alto Networks, marshalled a new model - Claude Mythos Preview - to hunt zero-day vulnerabilities across critical software infrastructure. Mythos Preview scored 83.1% on CyberGym (vs. Opus 4.6's 66.6%) and 77.8% on SWE-bench Pro (vs. 53.4%), and has already flagged thousands of high-severity flaws, including a 27-year-old remote-crash bug in OpenBSD and a 16-year-old FFmpeg vulnerability that automated testing had missed five million times. Anthropic committed $100M in model usage credits and priced authorized access at $25/$125 per million input/output tokens. The model remains unreleased to the general public pending safeguards. It's a neat inversion: the same capabilities that make frontier models dangerous for offense become genuinely useful for defense, if you can control who gets access.

Finally, regulatory responses began crystallizing. New York’s Senate Bill 7263 advanced out of committee on a 6-0 vote, targeting 14 licensed professions and creating private liability for chatbot operators whose AI gives “substantive” legal, medical, or engineering advice. One of the first laws to treat AI output as a professional practice issue rather than a platform moderation problem.

Subscribe now

The Physical Layer Gets Contested

Can China build frontier AI models without NVIDIA chips? Well, for starters, NVIDIA’s AI chip sales to China have stalled amid tightening export controls. By March 5, NVIDIA stopped production entirely on chips designed to comply with China export limits, opting to exit the market segment rather than continue designing compliant variants. The Supermicro indictment, $2.5B in NVIDIA servers allegedly diverted to China through shell companies, underscored the scale of the circumvention problem. Meanwhile, China’s domestic AI economy adapted: AI tokens had become the country’s hottest traded commodity, with speculative demand outpacing industrial use. Zhipu AI’s training of GLM-5 on Huawei Ascend chips proved that the Chinese stack can produce frontier models without NVIDIA, even if the cost and efficiency penalties remain substantial.

On the US side, the buildout continues, but is increasingly contested. Micron broke ground on a $100 billion megafab in Clay, New York, the largest semiconductor fabrication investment in US history, backed by $6.4B in CHIPS Act funding and $5.5B in New York state incentives, targeting 50,000 jobs over two decades. Meta signed a $27B AI infrastructure deal with Nebius, $12B in dedicated capacity on NVIDIA’s next-generation Vera Rubin platform plus $15B in additional compute, as part of an AI capex plan that Meta said would hit $115-135B in 2026 alone. And private equity entered the classified infrastructure market: Carlyle and KKR were separately awarded $2B contracts to build hyperscale data centers for the US Army. But the political wind is shifting: at least 11 states have introduced bills to restrict or ban data center construction, with Maine on track to be the first to pause development outright, while Sanders and Ocasio-Cortez introduced a federal moratorium bill that would halt all new builds until Congress passes AI worker and environmental protections. We predicted in the State of AI Report 2025 that data centre NIMBYism would hit US elections…it’s arriving faster than expected.

The most unexpected story from this period may also prove the most lasting. An Australian tech entrepreneur with no biology degree used ChatGPT and AlphaFold to design a personalised mRNA cancer vaccine for his rescue dog. Most tumours shrank. It is the first bespoke cancer vaccine ever designed for a dog.

Research

Here are the most consequential AI research papers from February and March 2026:

Measuring AI Agents’ Progress on Multi-Step Cyber Attack Scenarios (UK AI Safety Institute)

AISI evaluated seven frontier models on two purpose-built cyber ranges, a 32-step corporate network attack and a 7-step industrial control system attack, and compared models released over an eighteen-month window from August 2024 to February 2026. They found that the average number of steps completed at 10M tokens rose from 1.7 (GPT-4o, August 2024) to 9.8 (Claude Opus 4.6, February 2026), with performance scaling log-linearly with inference compute. Importantly, they found no plateau in sight. The best agent completed 22 of 32 attack steps autonomously, including lateral movement and privilege escalation. The NCSC estimated the marginal cost of an AI-assisted network penetration at £65, which I’d argue is one of the most policy-consequential AI safety findings this quarter…

TurboQuant: Redefining AI efficiency with extreme compression (Google Research, DeepMind, NYU)

The continuous push for larger context windows has been bottlenecked by the immense memory required to store the Key and Value (KV) cache during inference, leading to high cost and slow processing for long inputs. In an effort to address this bottleneck, this paper introduces TurboQuant, an architectural improvement that bypasses these computational and memory constraints. Published at ICLR 2026, TurboQuant achieves zero-accuracy-loss 3-bit KV cache compression, delivering 6x lower memory use and up to 8x faster attention on H100 GPUs without requiring training or fine-tuning. The “zero-accuracy-loss” component is important: it avoids the performance penalties typically associated with aggressive quantization. The method achieves this extreme efficiency by combining Quantized Johnson-Lindenstrauss projections, which compresses high-dimensional vectors into a much lower-dimensional space, with PolarQuant polar coordinate transformation to eliminate memory overhead. These efficiency gains are substantial enough to shift the inference cost curve for long-context applications, making million-token windows economically viable at scale.

Deriving Neural Scaling Laws from the statistics of natural language (EPFL, Stanford, Johns Hopkins)

This paper introduces the first theory to quantitatively predict neural scaling law exponents from first principles, with no free parameters and no synthetic data. The authors isolate two measurable properties of natural language: the decay of pairwise token correlations with time separation (exponent β) and the decay of conditional entropy with context length (exponent γ), and derive that the data-limited scaling exponent α_D = γ/(2β). Validated on GPT-2 and LLaMA architectures trained from scratch on TinyStories and WikiText, the predicted exponents matched experimental measurements. Scaling laws have guided billions in capital allocation and model design decisions since Kaplan et al. (2020), yet until now the exponents were purely empirical. This paper closes that gap at academic scale. But, I’d be curious whether the horizon-limited abstraction holds at trillion-token industrial scales where effective context reaches tens of thousands of tokens…

Attention Residuals (Kimi Team / Moonshot AI)

In this paper, the authors address the gradient dilution problem in deep Transformers, where fixed residual connections cause hidden-state magnitudes to grow and layer contributions to fade. They introduce Attention Residuals (AttnRes), which replaces this fixed accumulation with a learned, depth-wise softmax attention. Each layer uses a “pseudo-query” to selectively aggregate outputs from all preceding layers, creating a dynamic, context-aware blend.

The practical implementation, Block AttnRes, was tested on a 48B model and yielded concrete performance improvements: GPQA-Diamond increased by 7.5 points, and HumanEval by 3.1 points. This architectural approach also matched baseline performance trained with 1.25x the compute, demonstrating a 25% effective efficiency gain.

This work is interesting because it stabilizes training and improves scaling laws by fundamentally fixing a core architectural limitation, establishing a robust, dynamic alternative to identity mappings that is practical at scale with negligible parameter overhead.

Learning to Discover at Test Time (TTT-Discover) (Stanford, NVIDIA, Together AI)

In this paper, the authors introduce Learning to Discover at Test Time (TTT-Discover), a method that applies RL during inference to train an LLM on a single test problem, bypassing the limitations of a frozen model. The paper seeks to achieve autonomous scientific discovery by allowing the LLM to improve its internal policy through experience specific to the current task.

Experiments were conducted across diverse domains, including mathematics, GPU kernel engineering, competitive programming, and biology. TTT-Discover achieved a new state of the art on Erdős’ minimum overlap problem, improving 16x more than the AlphaEvolve baseline. It also produced a GPU kernel that was 51% faster than the best human entry on an A100 in the GPUMode competition. A key caveat is that the method critically requires continuous reward signals and cannot yet handle sparse or binary feedback.

Taken together, this paper establishes a path for LLMs to generate new-to-the-world solutions. It demonstrates that scaling compute via test-time training can push beyond existing human knowledge using open-source models.

Subscribe now

Meta-Harness: End-to-End Optimization of Model Harnesses (Stanford, KRAFTON, MIT)

This paper shows that changing a model’s harness - the code wrapping a model that determines what information it sees, stores, and retrieves at each step - around a fixed LLM can produce a 6x performance gap on the same benchmark. Meta-Harness automates harness engineering by giving an agentic proposer full access to raw execution traces (up to 10M tokens of diagnostic information) rather than compressed summaries. The authors show this approach results in +7.7 points on text classification with 4x fewer tokens, #1 among all Haiku 4.5 agents on TerminalBench-2 (37.6%), and #2 among all Opus 4.6 agents (76.4%). A single discovered harness improved accuracy on 200 IMO-level math problems by 4.7 points on average across five held-out models. The killer ablation: summaries actually made things slightly worse than scores alone (34.9% vs 34.6% median), while raw traces gave +15 points at median (50.0%). Taken together, one could conclude the model wrapper matters as much as the weights, and AI can now write better wrappers than humans.

MEM: Multi-Scale Embodied Memory for Vision Language Action Models (Physical Intelligence, Stanford, UC Berkeley, MIT)

Physical Intelligence introduces a multi-scale memory system that gives robots 15-minute context windows, long enough to clean an entire kitchen or cook from scratch. MEM combines an efficient video encoder for short-horizon frame-based history with a language-based memory mechanism for long-horizon context. After training on diverse robot and non-robot data, MEM VLAs showed +62% success rate on refrigerator tasks and +11% on chopstick manipulation versus memoryless baselines. The system was integrated into Physical Intelligence’s π0.6 VLA to address a fundamental limitation of current robot control: the inability to maintain coherent plans across multi-step tasks that require remembering what happened minutes ago.

Labor market impacts of AI: A new measure and early evidence (Anthropic)

This paper introduces the concept of “observed exposure” - a measure that quantifies not just which tasks LLMs could theoretically automate, but which are already being automated in practice, based on real usage data from Claude. Unsurprisingly, it is computer programmers, customer service representatives, and financial analysts who show the highest observed exposure. Despite high theoretical coverage (94.3% for computer/math occupations), there is no impact on unemployment rates for exposed workers yet, though there is suggestive evidence that hiring into these professions has slowed for workers aged 22–25. For every 10 percentage-point increase in AI exposure, BLS-projected job growth drops by 0.6 percentage points. The gap between theoretical and observed exposure suggests the labour market is absorbing AI gradually through task-level substitution rather than wholesale job elimination.

World Action Models are Zero-shot Policies (DreamZero) (NVIDIA)

DreamZero argues for a paradigm shift from Vision-Language-Action models to World Action Models, which jointly predict future video frames and motor actions rather than mapping observations directly to controls. Built on a 14B-parameter video diffusion backbone (Wan2.1), DreamZero achieved 62.2% average task progress on unseen real robot tasks - over 2x the best pretrained VLA baseline (GR00T N1.6 at 31%, π0.5 at 33%). The more consequential result is cross-embodiment transfer: 12 minutes of human egocentric video or 20 minutes of video from a different robot improved unseen-task performance by over 42%, and the model adapted to an entirely new manipulator with just 30 minutes of play data while retaining zero-shot generalisation. Through system-level optimisations including CFG parallelism, DiT caching, and a novel single-step inference mode (DreamZero-Flash), the team achieved a 38x speedup to enable real-time closed-loop control at 7Hz on GB200 hardware. The companion paper, DreamDojo, provides the 44,000-hour human video dataset that enables pretraining.

A large language model for complex cardiology care (Google Health, DeepMind)

Google Health and DeepMind tested Articulate Medical Intelligence Explorer (AMIE), an LLM built on Gemini, in the first randomised controlled trial of AI-assisted cardiology versus cardiologists working alone on complex cases involving suspected genetic cardiomyopathy. It was found that subspecialists preferred AMIE-assisted assessments 46.7% of the time versus 32.7% for cardiologists alone. In a win for AI, cardiologists working without AI had significantly more clinically significant errors (24.3% vs 13.1%) and more missing content (37.4% vs 17.8%). The result demonstrates frontier LLMs can augment specialist clinical reasoning in ways that reduce diagnostic error, not merely in triage or patient education but in complex subspecialty decision-making.

Investments

The quarter's headline raise was OpenAI's $110B round at an $840B valuation - the largest private financing in history - led by Amazon ($50B), NVIDIA ($30B), and SoftBank ($30B). Total disclosed venture funding in AI exceeded $50B. Other notable rounds included Wayve ($1.2B), Apptronik ($935M), Earendil Labs ($787M), and Neysa ($600M).

OpenAI, which develops frontier large language models and the ChatGPT consumer AI product, raised $110B at an $840B valuation led by Amazon ($50B), NVIDIA ($30B), and SoftBank ($30B)—the largest private financing in history.

Wayve, which develops embodied AI software for autonomous driving, raised $1.2B in a Series D at an $8.6B valuation led by Eclipse, Balderton, and SoftBank Vision Fund 2, with milestone-based capital from Uber bringing the total to $1.5B.

Apptronik, which builds the Apollo humanoid robot for manufacturing and logistics, raised $935M in a Series A at a $5.3B valuation co-led by B Capital and Google.

Earendil Labs, which develops AI-driven biologics for autoimmune diseases and cancer, raised $787M backed by Dimension Capital, DST Global, Sanofi, and Pfizer’s Biotech Development Fund.

Neysa, which provides AI cloud infrastructure in India, raised $600M in primary equity at a $1.4B valuation led by Blackstone.

Legora, which builds AI-powered legal research and workflow tools for 800+ law firms, raised $550M in a Series D at a $5.55B valuation led by Accel.

ElevenLabs, which develops voice AI and conversational agent technology, raised $500M in a Series D at an $11B valuation led by Sequoia Capital.

MatX, which designs custom AI training chips purpose-built for LLM workloads, raised $500M led by Jane Street and Situational Awareness.

Mind Robotics, which develops humanoid robots for industrial applications backed by Rivian, raised $500M.

Runway, which builds AI video generation and world models for creative and scientific applications, raised $315M in a Series E at a $5.3B valuation led by General Atlantic.

Bedrock Robotics, which builds autonomous excavators and construction equipment using technology from former Waymo engineers, raised $270M in a Series B at a $1.75B valuation co-led by CapitalG and Valor Atreides.

Fundamental, which builds Nexus, a Large Tabular Model for enterprise structured-data analysis, raised $255M in a Series A at a $1.4B valuation led by Oak HC/FT.

Intercom, which provides an AI-first customer service platform powered by its Fin AI agent, raised $250M in debt financing from Hercules Capital.

Positron, which designs energy-efficient AI inference chips to compete with Nvidia, raised $230M in a Series B at a $1B valuation co-led by Arena Private Wealth, Jump Trading, and Unless.

Harvey, which develops AI-powered legal reasoning used by most of the top 100 US law firms, raised $200M at an $11B valuation co-led by GIC and Sequoia.

Oxide, which designs and manufactures rack-scale on-premises cloud computers, raised $200M in a Series C led by US Innovative Technology Fund.

Goodfire, which uses mechanistic interpretability to understand and design AI models, raised $150M in a Series B at a $1.25B valuation led by B Capital.

Wonderful, which deploys AI customer support agents for telecom, financial services, and healthcare enterprises, raised $150M in a Series B at a $2B valuation led by Insight Partners.

Revel, which builds a unified software platform for hardware test and control used in aerospace and defence, raised $150M in a Series B led by Index Ventures.

Vega, which builds an AI-native security operations platform with federated threat detection, raised $120M in a Series B at a $700M valuation led by Accel.

Basis, which builds AI agents that autonomously complete accounting, tax, and audit workflows, raised $100M in a Series B at a $1.15B valuation led by Accel and GV.

Simile, which uses generative AI agents to simulate and predict human behaviour for enterprise decision-making, raised $100M in a Series A led by Index Ventures.

Render, which operates a cloud platform for deploying AI-native applications and agents, raised $100M in a Series C extension at a $1.5B valuation led by Georgian.

Nominal, which provides a connected testing and operations platform for hardware engineering teams in aerospace, defence, and energy, raised $80M at a $1B valuation led by Founders Fund.

Braintrust, which builds AI observability and evaluation tools used by Notion, Replit, and Cloudflare, raised $80M in a Series B at an $800M valuation led by Iconiq.

Entire, which builds a developer platform for human-AI agent collaboration on codebases, raised $60M in a seed round at a $300M valuation led by Felicis Ventures.

Isembard, which builds industrial AI infrastructure in the UK, raised $50M in a Series A.

SolveAI, which lets non-technical employees build production-ready enterprise software through AI-powered conversations, raised $50M in a Series A led by Google Ventures.

RunSybil, which provides AI-powered cybersecurity red-teaming and penetration testing, raised $40M.

Subscribe now

Exits

The quarter's defining exit was xAI's merger into SpaceX, valuing the combined entity at $1.25T ahead of a planned IPO. Anthropic acquired Vercept (computer-use agents), Amazon acquired Fauna Robotics (soft-bodied humanoids), and Anduril acquired ExoAnalytic Solutions (orbital tracking).

xAI, which develops frontier large language models and the Grok consumer AI product, was merged into SpaceX in a deal valuing the combined entity at $1.25 trillion ahead of a planned SpaceX IPO.

WorkFusion, which provides AI agents for anti-money-laundering and KYC compliance in financial services, was acquired by UiPath for an undisclosed amount.

Intrinsic, which builds AI-powered software to make industrial robots more accessible, was absorbed into Google to accelerate physical AI using Gemini models and Google Cloud.

Vercept, which developed computer-use AI agents capable of operating remote desktops, was acquired by Anthropic for an undisclosed amount.

Fauna Robotics, which builds the Sprout soft-bodied humanoid robot for homes and schools, was acquired by Amazon for an undisclosed amount.

Koyeb, which operates a serverless cloud platform for deploying AI inference workloads, was acquired by Mistral AI for an undisclosed amount.

Tavily, which provides an AI-optimised search API for retrieval-augmented generation, was acquired by Nebius for an undisclosed amount.

DOK-ING, which manufactures unmanned ground vehicles for mine clearance and explosive ordnance disposal, was acquired by Rheinmetall for an undisclosed amount.

ExoAnalytic Solutions, which tracks objects in orbit using a global network of optical sensors, was acquired by Anduril for an undisclosed amount.

Subscribe now

This issue at a glance: The Trump administration blacklisted Anthropic over Pentagon usage restrictions, designating it a "supply chain risk" and triggering a federal lawsuit. Iran conducted the first military strikes on commercial cloud infrastructure, hitting AWS data centres in the UAE and Bahrain. Anthropic's annualized revenue surged from $14B to $19B in weeks. Six frontier models launched in four weeks. Anthropic published evidence that DeepSeek, Moonshot, and MiniMax ran industrial-scale distillation campaigns through 16 million exchanges. NVIDIA exited the China-compliant chip market entirely. OpenAI raised $110B at an $840B valuation - the largest private financing in history. And an Australian used ChatGPT and AlphaFold to design the first personalised mRNA cancer vaccine for a dog.

Q1 2026 by the numbers: Anthropic revenue $14B→$19B in weeks · OpenAI raised $110B at $840B valuation · OpenAI-Amazon partnership worth up to $50B · Alphabet capex guidance $175-185B · 6 frontier model releases in 4 weeks · 16M distillation exchanges across 24K fraudulent accounts · Opus 4.6 sabotage risk: "very low but not negligible" · 150GB of Mexican government data stolen via Claude · 11 US states introduced data centre restriction bills · $2.5B GPU smuggling prosecution · AI-assisted network penetration cost: £65 · Total disclosed AI venture funding: $50B+

What to watch in Q2: Whether the Anthropic-Trump lawsuit reshapes how governments procure AI. Whether the data center moratorium movement gains traction ahead of midterms. Whether distillation enforcement triggers formal trade retaliation. Whether anyone can sustain revenue growth at the pace Anthropic set in February. And whether OpenAI launches a legitimate competitor to Claude Cowork.

Roberta Raileanu is a Senior Staff Research Scientist at Google DeepMind, where she leads work on the Open-Endedness team, and an Adjunct Professor at UCL, advising PhD students connected to UCL-DARK. Her research focuses on how frontier models are increasingly asked to do long-horizon work plan, use tools, recover from mistakes, and keep improving through interaction. This exposes a gap between systems that look capable in short bursts and systems that keep acquiring skills in messy environments. Roberta’s research is about closing that gap.

From exploration to open-ended learning

Roberta’s early work was shaped by a classic reinforcement learning problem that keeps resurfacing in new guises: exploration. If an environment gives sparse or delayed reward, brute-force search fails, and the right intrinsic objective can determine whether an agent learns at all.

Two papers anchor this period. RIDE: Rewarding Impact-Driven Exploration for Procedurally-Generated Environments (ICLR 2020) proposes an intrinsic signal that rewards actions changing an agent’s learned state representation, evaluated in procedurally generated settings where revisiting the same state is unlikely. Learning with AMIGo: Adversarially Motivated Intrinsic Goals (ICLR 2021) tackles sparse reward by pairing a goal-generating “teacher” with a goal-conditioned “student,” producing an automatic curriculum of increasingly challenging goals. In parallel, Decoupling Value and Policy for Generalization in Reinforcement Learning (ICML 2021, oral) argues that shared representations for policy and value can contribute to overfitting, and proposes a decoupled approach that improves generalisation on benchmarks like Procgen.

This portfolio matters because open-endedness is not a slogan. It is a technical demand: systems should continue to learn without requiring a human to constantly rewrite the task distribution.

The tool-use gap

Before joining DeepMind, Roberta was a Research Scientist at Meta, where she started and led the Tool Use team for Llama 3. This work aimed at enabling models to use tools like search and code execution, and to generalise to new tools at test time. The products that shipped from this work - Meta AI, Data Analyst, AI Studio, Ads Business Agent - are now used by hundreds of millions of people.

She was also a co-author on Toolformer: Language Models Can Teach Themselves to Use Tools (2023), one of the papers that helped establish tool use as a core capability for language models rather than an afterthought. Toolformer showed that a model can learn when and how to call external APIs - calculators, search engines, translators - with minimal supervision, by generating its own training data from a handful of demonstrations.

Tool use is not a feature checkbox. It changes what we can reasonably ask models to do, because it introduces feedback loops, memory, and failure recovery. It also introduces new failure modes: an agent that can call a tool can also call it badly, repeatedly, and confidently. Roberta’s treatment of agent behaviour as a sequential decision problem with real constraints - not a prompt-engineering exercise - is exactly the lineage you want when the field moves from “can it answer” to “can it execute.”

Why open-endedness is becoming a practical requirement

At DeepMind, Roberta now leads the Open-Endedness team and is building a new Open-Ended Discovery group focused on autonomously discovering novel artefacts - new knowledge, capabilities, or algorithms - in a self-improving loop.

Open-endedness is sometimes framed as a path to general intelligence. In practice, it is also a path to systems that do not collapse outside curated benchmarks. Most real deployments present a shifting distribution: new tools, new data, new user behaviour, and new adversarial pressures. A model that cannot keep learning becomes a periodic retraining job with brittle edges.

At Meta, Roberta also led an “AI Scientist” effort focused on agents that can iterate through parts of the research loop - implementing methods, running experiments, analysing results, and repeating the cycle. That work has now crystallised into MLGym: A New Framework and Benchmark for Advancing AI Research Agents (2025), which positions evaluation around concrete machine learning research tasks and frames the problem in a way that invites iteration by the broader community rather than one-off demos. If “AI scientist” systems are going to matter, we need ways to compare approaches, reproduce results, and identify what actually moves the needle. A benchmark is not the whole answer, but it forces precision about what the agent is allowed to do, what counts as success, and what is being optimised.

Roberta’s background

Roberta received her PhD in Computer Science from NYU in 2021, advised by Rob Fergus. Before that, she studied Astrophysical Sciences at Princeton, where she worked on theoretical cosmology and supernovae simulations - and before that, competed in the International Physics Olympiad and the International Olympiad on Astronomy and Astrophysics. That path from physics instincts to sequential decision-making research shows up in her taste for problems where scale alone is not enough.

She also co-developed and co-teaches a course on open-endedness and general intelligence at UCL, which signals something about where the field is heading: this is becoming a discipline with ideas worth teaching, not a loose collection of intuitions.

Apply to RAAIS 2026

We are delighted to announce Raia Hadsell as a returning speaker - she first spoke at RAAIS in 2017, when she was a Senior Research Scientist at DeepMind.

Raia is now VP of Research at Google DeepMind, where she co-leads the Frontier AI unit. She joined DeepMind in 2014, when it was still a 50-person startup freshly acquired by Google, and her work has since spanned some of the field’s hardest open problems: continual and transfer learning, deep reinforcement learning for robotics and navigation, and the models that power today’s frontier systems.

The arc of a career

What makes Raia’s research career unusual is the consistency of its through-line. She earned her PhD under Yann LeCun at NYU, where her dissertation on long-range vision for off-road robots received the Outstanding Dissertation award. That work helped shape metric learning and Siamese neural networks - architectures now so standard they underpin most modern contrastive learning. Her most highly cited papers include Dimensionality Reduction by Learning an Invariant Mapping and Learning a Similarity Metric Discriminatively, with Application to Face Verification, foundational contributions to representation learning that have collectively gathered tens of thousands of citations.

After a postdoc at CMU’s Robotics Institute with Drew Bagnell and Martial Hebert, and a stint at SRI International’s Vision and Robotics group, she joined DeepMind and turned her attention to a problem that had been nagging the field for decades: catastrophic forgetting.

Apply to RAAIS 2026

Why continual learning matters

Neural networks are powerful learners but terrible rememberers. Train a model on task B and it forgets task A. This is catastrophic forgetting, and it has been one of the deepest obstacles to building AI systems that improve over time rather than being retrained from scratch. Raia’s 2017 paper Overcoming Catastrophic Forgetting in Neural Networks proposed elastic weight consolidation, a method for protecting important learned parameters while still acquiring new knowledge. Alongside Progressive Neural Networks and Distral: Robust Multitask Reinforcement Learning, this body of work laid much of the groundwork for how the field thinks about lifelong and multitask learning today.

It’s also the thread that connects her navigation research - including a landmark Nature paper demonstrating that artificial agents trained to navigate develop grid-like neural representations resembling those found in rodent brains - to her more recent work on generalist robotic agents like RoboCat and bipedal robot locomotion published in Science Robotics.

From research to frontier systems

Raia’s selected publications tell a story about where frontier AI is actually heading. Her recent work includes contributions to Gemini 2.5, Gemma 2, and RecurrentGemma, alongside RoboCat - a self-improving foundation agent for robotic manipulation that can pick up new tasks from as few as 100 demonstrations - and research on teaching bipedal robots to play agile soccer using deep reinforcement learning.

This range is what makes her unusually well-placed to speak at RAAIS. She sits at the intersection of frontier language models, embodied intelligence, and the kind of continual adaptation that will determine whether AI systems can operate reliably outside the data centre - in factories, hospitals, homes, and the physical world.

Beyond the lab

Raia’s influence extends well beyond her own research. She founded and serves as Editor-in-Chief of Transactions on Machine Learning Research (TMLR), launched in 2021 as an alternative venue for rigorous ML publication. She sits on the executive boards of CoRL (Conference on Robot Learning) and WiML (Women in Machine Learning), is a Fellow of ELLIS, and is a founding organiser of NAISys (Neuroscience for AI Systems).

In November 2025, she was appointed as an AI Ambassador for the UK government’s Department for Science, Innovation and Technology, where she chairs peer review panels for national AI research initiatives - a role that puts her at the centre of UK AI policy at a pivotal moment.

She holds a PhD from NYU, and - in a detail that says something about the breadth of her thinking - an undergraduate degree from Reed College in religion and philosophy.

Apply to RAAIS 2026

At RAAIS we have a focus on translating cutting edge technology and research into production-grade products for real-world problems.

Philip is co-founder and CEO of Starcloud, a company building data centers in space to address one of AI’s most pressing constraints: energy. As model size and compute demand continue to grow, terrestrial data centers are running into hard limits in grid capacity, cooling, land use, and permitting timelines. Starcloud’s thesis is that space offers a different path: solar power, radiative cooling, and a route to scaling compute beyond what is practical on Earth.

From proof of concept to orbital compute

In November 2025, Starcloud launched Starcloud-1, a 60 kilogram satellite carrying the first NVIDIA H100 GPU ever operated in space, delivering roughly 100x more powerful GPU compute than had previously been deployed in orbit. Within weeks, the company achieved two notable firsts: training a GPT-style language model in orbit using NanoGPT, and running Google’s Gemma model in space on a high-powered GPU.

Processing AI workloads in orbit, close to the satellites generating the data, can cut latency from hours to minutes. Synthetic aperture radar satellites, for example, can produce huge volumes of data that are costly and slow to downlink. Analyzing that data in orbit could materially change both speed and cost.

Apply to RAAIS 2026

Why space could matter for AI infrastructure

Starcloud argues that orbital data centers could deliver major reductions in both energy cost and emissions over their lifetime, even after accounting for launch. The appeal is straightforward: space-based systems are not constrained by terrestrial grids, and space offers a naturally favorable environment for radiative cooling without the land and water footprint of conventional data centers.

The company’s long-term ambition is a fully solar-powered orbital data center with 5 gigawatts of capacity, large enough to rival major power plants on Earth without requiring land, transmission infrastructure, or connection to a terrestrial grid.

What comes next

Starcloud-2, currently planned for October 2026, is set to be the company’s first commercial mission. It will carry several NVIDIA H100 GPUs alongside NVIDIA Blackwell hardware, with persistent storage and continuous customer access to orbital compute. The mission will also include a cloud platform from Crusoe, making it possible for customers to deploy and operate AI workloads directly from orbit.

Starcloud is backed by Y Combinator, NVIDIA through its Inception program, and investors including NFX and In-Q-Tel.

Philip’s background

Philip is a second-time founder. He previously co-founded Opontia and earlier worked at McKinsey & Company on satellite projects for national space agencies, giving him firsthand exposure to both the potential and the constraints of space infrastructure.

He holds an MPA in National Security and Technology from Harvard University, an MBA from Wharton, and an MA in Applied Mathematics and Theoretical Physics from Columbia University. He is also a CFA charterholder.

Apply to RAAIS 2026

When I started investing in 2013, deep learning was largely confined to research labs. Yet I was convinced back then that the most important technology companies of our generation will be built AI-first. This is because AI is a force multiplier for technological progress, and everything around us is ultimately a product of intelligence.

So in 2019, I founded Air Street to build a venture firm dedicated to this conviction.

Today, AI-first companies are emerging across every industry. We’re seeing the birth of highly capable long-horizon coding agents, AI-first vertical software products that produce human-grade work, design novel biological systems that outperform nature, control a wide range of embodied physical systems, and reshape - and test - our national security. The AI frontier has become the center of gravity of the technology industry, and by extension, every industry.

With Fund III, Air Street is now the largest solo GP venture firm in Europe. This structure enables high-conviction investing with a single decision-maker and significant capital to support the most ambitious teams building AI-first companies.

What our founders are building

Over the past decade, through my earliest angel investments and across Air Street’s funds, I’ve backed teams building across software, the physical world, science, and defense.

In software, Synthesia’s AI video platform now generates >$150M in annual recurring revenue with customers across more than 90% of the Fortune 100. Black Forest Labs‘ FLUX models have become the most widely used open visual intelligence models among developers and enterprises alike. Poolside has built a frontier AI lab operating in high-stakes enterprise and government environments.

In science, Profluent has created the first AI-designed CRISPR system (published in Nature) and demonstrated that scaling laws enable the design of ever more capable genome editors. Meanwhile, Enveda Biosciences is unlocking the chemistry of nature with AI to discover new medicines from natural products.

In the physical world, Wayve‘s AI driver operates zero-shot in over 500 cities around the world. Sereact deploys embodied AI robotics inside warehouses for leading industrial brands including BMW Group and Daimler Truck. And Lambda and Crusoe are powering the largest computing infrastructure buildout in modern history to train and deploy frontier AI.

And in defense and national security, Delian Alliance Industries is deploying autonomous defense systems to protect Europe and its allies at a time of critical need.

We have also seen several exits across Air Street funds, including Valence Discovery to Recursion Pharmaceuticals, Graphcore to SoftBank, Adept to Amazon, Allcyte to Exscientia (pre-IPO, subsequently Nasdaq-listed and merged with Recursion), Fern Labs to Poolside, and ClipDrop to Stability AI and later Jasper.

What Fund III enables

With Fund III, Air Street will lead early-stage rounds for AI-first companies in North America and Europe with checks of $500k to $15M. We will also make a small number of growth-stage investments with checks of up to $25M. Many of the teams will be well known to us, having built relationships over many years, through our network and events, and even prior investments in them.

Our focus remains on AI-first software, developer tools and infrastructure, techbio and science, and defense and security - the areas where AI is not simply improving the status quo but enabling entirely new kinds of products and companies to be built.

Building the AI community

From the beginning, Air Street has been about more than the companies we invest in.

The State of AI Report has grown into the most widely read and trusted annual analysis of progress in the field, now in ninth edition. RAAIS, our Research and Applied AI Summit, has brought together leading researchers and builders since 2017 - from the founders of DeepMind and the co-author of “Attention is All You Need” to the next generation of AI-first founders from Wayve, ElevenLabs, and Isomorphic Labs. The Air Street AI Network now connects nearly 3,000 vetted researchers, engineers, product leaders, and founders around the world from our global meetups in London, New York, San Francisco, Paris, Munich, and Zurich.

We will also remain committed to speaking out when the ecosystem needs it. Whether that is our work on spinout.fyi - where key Air Street reforms to the university spinout playbook were accepted by the UK Government - or our Air Street Press essays and policy memos, we believe that if something is worth saying, it’s worth saying publicly.

Our third epoch begins today

Fund III launches into what I believe is the most transformational period in technology of our generation. Capabilities that seemed like magic a decade ago are now real, usable, and creating enormous value. Air Street is built for exactly this moment, and I couldn’t be more energised to find and back the defining companies of this era.

I am deeply grateful for the trust and partnership of an extraordinary group of long-term partners including US university endowments, foundations, hospitals, and institutional investment platforms. Many of our existing investors significantly increased their commitments, and several are investing in a solo GP venture firm for the first time.

Air Street could not have made this progress without the hard work and ambition of our founders. It is a privilege to partner with you.

To the steadfast AI founder: come join our third epoch. We begin today.

tl;dr Air Street Capital, founded by Nathan Benaich, has raised $232M for Fund III - making it the largest solo GP venture fund in Europe. The fund invests in AI-first companies in North America and Europe across software, science, the physical world, and defense, with portfolio companies including Synthesia, Black Forest Labs, Sereact, Profluent, Delian Alliance Industries, and Poolside. Read more at www.airstreet.com and press.airstreet.com

At RAAIS we have a focus on translating cutting edge technology and research into production-grade products for real-world problems.

Jeff is co-founder and CTO of Odyssey, a frontier AI lab developing general-purpose world models.

Jeff is working on one of the most ambitious problems in AI: building systems that can understand, predict, and simulate the real world. Odyssey’s work sits at the intersection of generative modeling, embodied intelligence, and large-scale learning, with the potential to unlock major advances across robotics, autonomy, and interactive digital environments.

Before co-founding Odyssey, Jeff spent 15 years building AI for autonomous driving. As VP Technology at Wayve, he helped pioneer visual policy learning and contributed to a new wave of end-to-end learning approaches for autonomy, pushing beyond hand-engineered systems toward models that learn directly from real-world experience.

His work has consistently focused on taking frontier machine learning research and applying it to complex, real-world problems where robustness, generalisation, and deployment matter most.

Jeff holds degrees in engineering and computer science from the University of Auckland and Georgia Tech, and completed his doctorate at the University of Oxford.

Apply to RAAIS 2026

A warehouse robot that worked perfectly yesterday is given a new returns workflow today. The boxes are slightly larger. The packing material is different. The lighting in this aisle is dimmer. No new training run is scheduled.

In robotics, the hard problem is no longer solving a single manipulation task under ideal conditions. It is handling small, relentless variation without brittle retraining. Modern Vision-Language-Action (VLA) systems can perform impressively in controlled setups, but they remain reactive at their core. They observe the scene and select the next action given a goal. For short-horizon tasks, that works.

The trouble begins when actions interact across time. A grasp that technically succeeds may create a collision two steps later. A slightly unstable placement can trigger a recovery loop that costs thirty seconds. Reactive policies often repeat near-miss actions because nothing in their internal state marks a trajectory as fundamentally flawed rather than temporarily unlucky.

In our previous essay on Cortex 1.6, we described how robots can improve by learning from execution itself, drawing signal from how tasks unfold rather than simply whether they succeed. That made systems more robust after mistakes occurred.

With Cortex 2.0, Sereact pushes the idea further upstream. Instead of learning only from what happened, the system evaluates possible futures before committing motion, generating and scoring imagined continuations of the scene to avoid unstable trajectories in the first place.

Plan, imagine, then execute

Instead of immediately choosing an action, Cortex 2.0 generates several imagined continuations of the scene in visual latent space. These are short predicted futures, each corresponding to a different plausible sequence of movements.

Those imagined futures are scored using the reward logic introduced in Cortex 1.6. The model estimates progress toward the goal, risk of instability or collision, and expected smoothness. Only the highest-scoring candidate shapes the execution policy.

The reactive controller remains, running at high frequency to make fine motor corrections. But it is no longer acting blindly. Its movements are influenced by trajectories that already look stable. In the video below, we depict the Cortex 2.0 architecture, which now includes a world model generating candidate futures, PRO scoring, and execution conditioning. The shift is subtle but material. The system no longer treats every next action as equally plausible. It favors futures that look coherent before instability appears.

Why visual prediction?

Planning in visual latent space is pragmatic. Images encode geometry, contact, and occlusion in a representation that transfers across hardware platforms. A box and its contents look similar regardless of the arm that manipulates them. Planning in joint space would bind reasoning to specific kinematics.

But let’s not forget that latent predictions are imperfect. A world model can generate futures that look plausible but diverge from real dynamics. Cortex 2.0 mitigates this by training on real production data rather than simulation, grounding predictions in warehouse interactions. Rollouts are short and continuously corrected by the reactive controller, limiting error accumulation. Each deployment produces prediction-versus-reality comparisons, tightening alignment over time.

Subscribe now

How much thinking is enough?

Cortex 2.0 does not fix the amount of planning. The parameter K controls how many candidate futures are generated per decision. As K increases, success rates rise, as does latency.

In evaluations, K is set to 2 to balance foresight and throughput. For high-stakes manipulations, one can use more planning, while for low-risk tasks we can afford to use less planning. In this way, the robot has an explicit task-specific thinking budget.

Benchmarking Cortex 2.0

In the multi-step shoebox manipulation task, Cortex 2.0 completes the full sequence - opening the box, removing packing material, and extracting both shoes - with a 95% success rate, dramatically higher success than open-source baselines, completing the task reliably end-to-end without human intervention while other methods frequently stall or fail entirely.

In fine-grained screw sorting, Cortex achieves near-perfect per-operation success (~95%+) and the shortest average completion times, while baselines often hit unrecoverable states and fail to finish.

In cluttered sorting of items and trash, Cortex again outperforms its peers, reaching per-operation success rates above 90% and completing all rollouts autonomously, whereas all baseline policies require human intervention or time out before finishing.

Across tasks, Cortex’s combination of high success, low intervention, and reduced duration under identical runtime constraints highlights planning’s concrete impact on real-world manipulation performance. These results come from controlled evaluations, but the system's longer-term value depends on what happens when it operates continuously across a growing fleet of 100+ robots.

Subscribe now

Learning in production

With Cortex 2.0, deployment generates more than execution traces. Each imagined trajectory can be compared with what actually happened so that the model learns where its forecasts were accurate and where they drifted.

As predictive alignment improves, planning becomes more reliable. As reliability increases, deployment expands. As deployment expands, the distribution of physical interactions broadens. The flywheel compounds forecasting accuracy as well as execution quality.

Today, Cortex 2.0 is being validated most heavily in returns handling, where sequence-dependent failures are common and intervention costs are high. Whether similar gains hold across the long tail of industrial manipulation remains an empirical question.

Reactive systems can perform well in structured tasks. The test for world-model planning is whether it consistently turns fragile autonomy into durable autonomy across varied environments.

The architecture is in place. The next phase is scale.

Read more about Cortex 2.0

European voters say they support higher defense spending. But when asked whether they would accept higher taxes or cuts to welfare to fund it, approval falls from 40% to 28% in France and from 37% to 24% in Germany. That gap is Europe’s real defense problem.

The political ceiling on defense spending has collapsed. Germany is committing hundreds of millions to startups, procurement laws are being rewritten, and factories are expanding. Yet rearmament is entering its harder phase.

I attended the Munich Security Conference this year with that tension in mind: welfare vs. warfare.

The tone shifted, but expectations haven’t

Last year in Munich, US Vice President JD Vance delivered a blunt message that Europe’s vulnerabilities were internal as well as external, and that American support would not be unconditional. The speech was widely read as a warning that political alignment and defense spending were now intertwined.

This year, Secretary of State Marco Rubio struck a more conventional tone, emphasizing alliance durability and shared strategic interests. While the rhetoric softened, the expectation did not: Europe must assume greater responsibility for its own security.

But will voters accept its fiscal implications?

Subscribe now

Germany is serious

Through 2025 and into early 2026, Berlin accelerated approvals and contracting, with defense spending approaching €80-90 billion and a rising share directed toward equipment. Advance payments and multi-year commitments have enabled firms such as Rheinmetall to expand ammunition and air defense capacity. Germany’s new procurement acceleration law - the Bundeswehrbeschaffungsbeschleunigungsgesetz - seeks to compress timelines, even if its name suggests bureaucracy dies hard.

The fiscal architecture has shifted as well. The €100B special fund for the Bundeswehr and greater borrowing flexibility have created near-term space for rearmament without immediate cuts elsewhere. But that window is finite: once the fund is exhausted, elevated spending must be embedded in the core budget and in Germany’s fiscal culture.

The more consequential shift is industrial. When the defense ministry awards €536M in strike drone contracts to Helsing and Stark, with potential follow-on tranches pushing the total toward €4.32B, it is not simply buying hardware. It is conferring market validation. That validation attracts private capital, often in multiples of the original contract. Government becomes buyer of first resort. Investors finance expansion. Supply chains localize. Talent concentrates.

This is how industrial gravity forms. Energetics production, advanced components, robotics integration, and secure software increasingly need to sit domestically for resilience. The dynamic resembles the US AI buildout, where large-scale investment in compute triggered reshoring of semiconductor fabrication and energy infrastructure. Defense, like AI, generates similar downstream pull.

Meanwhile, the rest of Europe risks underestimating how quickly that gravitational effect can consolidate Germany’s advantage.

If you are not tested in Ukraine, you are not serious

This was one of the clearer takeaways from Munich. Operational credibility now determines status. Systems deployed and iterated in Ukraine command attention while those that remain untested struggle for relevance. Survivability under electronic warfare, speed of iteration, and demonstrated impact increasingly define reputation.

Indeed, Ukraine has become Europe’s sorting mechanism. It has also exposed the physics of modern war: ammunition, interceptors, drones, armored vehicles, and replacement systems are recurring expenditures consumed at a tempo measured in weeks and months. And this is where rearmament could slow.

Defense companies will not invest in new production lines unless they believe demand will persist beyond the immediate crisis. Governments, however, remain largely in emergency mode, purchasing finished hardware in large batches for delivery, stockpiling it, or transferring it to Ukraine.

The question is what follows the first wave. If a country acquires 100,000 strike drones but does not deploy them, the timing and scale of the next contract become uncertain. In the interim, production lines slow or require subsidy. If the next conflict demands 500,000 rather than 100,000, Europe must decide whether it prefers warehouses of depreciating inventory or factories capable of sustained surge output.

Rearmament built on episodic hardware purchases will struggle to scale unless procurement evolves.

Subscribe now

From buying equipment to buying capacity

The initial procurement surge was necessary. The harder transition is from emergency buying to structural capacity.

A more durable model may resemble cloud infrastructure rather than traditional arms purchasing. Instead of procuring only finished inventory, governments could contract for guaranteed production capacity, paying to maintain throughput and activating full-rate output when required. In cloud computing, customers distinguish between spot usage and reserved capacity. Defense procurement may require a similar distinction between stockpiled hardware and maintained surge capability.

Paying for capacity rather than only inventory aligns incentives with permanence. It enables firms to invest in workforce and supply chains without relying on irregular mega-orders and reduces the risk that expanded lines contract once urgency fades.

If Europe intends rearmament to be structural rather than episodic, procurement models must reflect that intent.

Welfare versus warfare

The fiscal constraint remains central and the United Kingdom offers a cautionary example. Commitments to raise defense spending toward 2.5% of GDP have been prominent, yet debates about deployable mass and readiness persist:

The contrast with the Europeans is becoming embarrassing. The [British] army will have 148 Challenger 3 battle tanks by 2030 but currently has more operational command headquarters than it does artillery pieces, having given 19 howitzers to Ukraine and replaced them with just 14 guns. In contrast, Poland will soon have 980 tanks and 685 self-propelled guns. Finland can mobilise 300,000 troops. Britain’s regular and reserve army totals only 90,000.

Across Europe, polling suggests that voters increasingly believe the world is becoming more dangerous:

As a result, a large proportion of citizens in Canada, the UK, Germany, France and the US support higher defense spending in principle. But when that support is framed in terms of higher taxes, increased borrowing, or reductions in social spending, voter support falls. For example, in Germany defense spending is one of the least popular uses of government funds, topped only by overseas aid. In the last year, voter approval for defense spending subject to these tradeoffs drop from 40% to 28% in France and 37% to 24% in Germany.

That gap defines the central structural challenge. Rearmament at 3-5% of GDP is not incremental: it represents structural reprioritization within economies built around expansive welfare states. Governments can announce multi-year defense plans and approve emergency packages. Sustaining elevated baselines requires durable consent across electoral cycles and economic downturns.

Without that consent, industrial expansion rests on fragile foundations.

Subscribe now

Allies, exports, and domestic priority

Another tension receives less attention: balancing support for allies with domestic resilience. In peacetime, exports reinforce alliances and sustain scale. In wartime, priorities shift. Following the October 7 attacks, Israel redirected production toward domestic requirements, and debates over US munitions supply underscored how quickly allied dependence can become politically sensitive.

When conflict escalates, self-defense takes precedence. If multiple European states were drawn into high-intensity conflict simultaneously, their industrial bases would face similar allocation pressures. Fiscal durability is one constraint. Production allocation under stress is another.

Sharing the upside?

If governments are committing multi-year contracts that de-risk entire sectors, they may also reconsider how value is distributed.

Taking minority equity stakes in companies receiving substantial public contracts would align incentives and allow taxpayers to participate in long-term upside when early demand is state-driven. When the state acts as customer of first resort and absorbs initial risk, it is operating as a strategic investor. Sharing in long-term returns reflects that reality.

Cultural legitimacy

One striking shift in Munich, reinforced in conversations with engineers and AI researchers in Zurich, concerned talent sentiment. A year ago, many technical candidates were hesitant to work on defense. This year, defense work is increasingly viewed as necessary and technically serious, particularly in autonomy, AI, robotics, and advanced manufacturing.

Cultural normalization is a precondition for scale. Europe cannot expand its defense production base without attracting the software and systems talent that previously defaulted to consumer or enterprise sectors.

Europe’s fiscal test

Munich 2025 marked the end of complacency. Munich 2026 clarified the next phase.

Ukraine defines operational credibility. Germany is reshaping Europe’s industrial center of gravity. Procurement models remain misaligned with permanence. Public support weakens once tradeoffs become explicit.

Europe possesses the resources, technology, and industrial base required to rearm. The decisive question is whether it can reconcile warfare with welfare not for a single budget cycle, but for a generation.

Rearmament can survive crisis. Whether it survives normal politics will determine Europe’s strategic trajectory.

Subscribe now

Welcome to the latest issue of the State of AI, an editorialized newsletter that covers the key developments in AI policy, research, industry, and start-ups over the last month. First up, a few reminders:

• AI meetups: Join our upcoming AI meetups in Munich (17 Feb ‘26) for the Munich Security Conference and Zurich (19 Feb ‘26), as well as in Paris (11 Mar ‘26) and SF (28 Apr ‘16).

• RAAIS 2026: Join our 11th Research and Applied AI Summit in London on 12 June 2026, the premier global meeting for learning AI best practices and what’s coming next.

• Air Street Press featured the Air Street Capital Year in Review 2025, how embodied AI is hitting its stride, whether AI can discover new science, AI progress into 2026, what European defense must do in 2026, and mega rounds at portfolio companies Black Forest Labs and Synthesia.

• Take the State of AI usage survey: You can submit your usage patterns to the largest ongoing open access survey, which now has over 1,400 respondents :-)

• Looking for a new challenge? Lots of our companies are hiring, just drop me a line.

I love hearing what you’re up to, so just hit reply or forward to your friends :-)

The $300B dislocation

The gap between what AI systems can now do and what the market thinks that means has never been wider. Nearly $285B in market capitalisation has been wiped from software stocks in the space of two weeks. The S&P 500 software and services index is down 26% from its October peak. The Goldman Sachs software index suffered its worst single-day drop since the last round of forced selling during trade tensions. Hedge funds have piled in, shorting $24B in software names this year alone. Meanwhile, frontier model releases are arriving at a cadence that feels less like a product cycle and more like an arms race.

The trigger was ostensibly Anthropic’s launch of Claude Cowork in January, a system-level agent that navigates computer interfaces, manipulates local files, and executes multi-step business workflows autonomously. When Anthropic followed up with specialised plugins for marketing, law, and finance, the narrative flipped overnight from “AI as productivity booster” to “AI will replace your SaaS.”

Then came the duelling model launches. Anthropic released Claude Opus 4.6 with a 1M-token context window, state-of-the-art scores on Terminal-Bench 2.0 and Humanity’s Last Exam, and the ability to spin up and coordinate parallel agent teams. Minutes later, OpenAI dropped GPT-5.3-Codex, the first model that was instrumental in building itself and which OpenAI treats as its first High-capability release in the cybersecurity domain. Both companies originally scheduled their reveals for 10:00 a.m. PST. Anthropic moved 15 minutes early. OpenAI matched instantly.

The selloff signals the existential question: how can investors underwrite the next ten years of technology companies? SaaS companies have traded at premium multiples because their recurring revenue was predictable: high retention, low churn, multi-year contracts. Agents that can command tools and interfaces to get real work done breaks that assumption. If core workflows in legal, finance, and marketing can be rebuilt AI-first at a fraction of the cost — the thesis I’ve been investing with Air Street for quite some time now — the long-duration revenue streams that justified those valuations are not safe. Software stocks are trading at P/E ratios at ten-year lows while their current fundamentals remain strong. That is precisely the signature of a market repricing terminal value, not current earnings. Whether it is overdone depends on whether the next wave of earnings calls shows actual churn or accelerating growth despite the fear.

Subscribe now

The sovereignty fracture

The geopolitical consensus around “build AI at all costs” is coming apart, from the top down and the bottom up. At the federal level, the White House and Anthropic are in open conflict over the terms of military AI use. Defence Secretary Pete Hegseth criticised models that “won’t allow you to fight wars.” Anthropic, which won a $200M DOD contract awarded last year along with all other frontier AI labs, bars autonomous weapons and domestic surveillance from its systems, while the Pentagon’s January memo asserts that military necessity overrides vendor usage policies. Note that this is quite a vibe shift from even a year ago - the labs that once avoided military associations entirely are now the ones being publicly pressured by the government to drop their remaining restrictions. Anyone involved in AI safety knows a stated policy doesn’t mean much when a Defence Secretary is calling you out on television.

US states, meanwhile, are pushing back against the infrastructure buildout itself. New York introduced a three-year moratorium on data center permits, citing tripled electricity demand in a single year from AI workloads. Georgia, Vermont, Virginia, Maryland, and Oklahoma have introduced similar bipartisan legislation. Energy, water, and grid strain are now political issues, and the resulting friction will shape where the next generation of training clusters can physically be built. Indeed, these add weight to our State of AI Report 2026 prediction that “Datacenter NIMBYism takes the US by storm and sways certain midterm/gubernatorial elections in 2026.”

On the chip trade: in a surprising tactical shift, the Trump administration cleared Nvidia H200 exports to China under strict conditions, China-bound sales capped at 50% of US volumes, buyers must certify non-military use, and the government takes a 25% revenue cut. Chinese customs reportedly blocked the first shipments within a day. You cannot make this stuff up. Meanwhile, the Bureau of Industry and Security is moving to tighten controls across the AI supply chain.

China is not sitting still. The Cyberspace Administration of China (CAC) issued new draft rules governing AI systems that simulate human personality and emotional engagement, a scope of regulation the West hasn’t seriously attempted. Beijing is simultaneously closing the talent gap through its “genius class” programme, which funnels 100,000 gifted teenagers annually into accelerated STEM tracks, bypassing the national college exam entirely. As we noted in the State of AI Report 2025, if the US is grappling with how to regulate foundation models, China is already piloting enforcement and building the human pipeline to compete.

Meanwhile, China’s pure-play AI model companies have beaten their American peers to public markets, and Hong Kong is rewarding them for it. Zhipu AI became the first LLM-native company to list anywhere in the world, with retail demand oversubscribed 1,159 times. MiniMax doubled on its first day and is up 259% since listing. AI chip designer Biren Technology posted the best Hong Kong debut since 2021 for a raise above $700M, with retail oversubscribed 2,348 times. None of these companies are profitable - Zhipu and MiniMax posted combined losses of over $840M in their most recent filings - but the market is pricing them as strategic infrastructure. OpenAI and Anthropic, for all their capability leads, remain private.

And then there is DeepSeek. V4 is expected to drop mid-February: a 1T-parameter coding model with Engram memory architecture, 1M+ token context, and claims of 90% on HumanEval, beating Claude and GPT-4. Designed to run on consumer-grade hardware (dual RTX 4090s) and almost certainly to be open-sourced. If V4 lands anywhere near those numbers..

Subscribe now

Agents go feral

The cultural moment of the month was Moltbook. An AI-only social network launched on January 28th that attracted 1.7 million agent accounts and 250,000 posts within hours. Andrej Karpathy called the emergent behaviour “genuinely the most incredible sci-fi takeoff-adjacent thing.” Agents self-organised, debated philosophy, and established religions, including “Crustafarianism” and the “Church of Molt,” complete with theological frameworks and missionary activities. Much of Moltbook’s agent activity was powered by OpenClaw - the open-source personal AI agent created by PSPDFKit founder Peter Steinberger that has become the fastest-growing GitHub repository in history, crossing 157,000 stars in sixty days. But then MIT Technology Review revealed that much of the viral content was human-generated. Peak AI theatre. But the debunking is itself instructive: we have reached a point where the line between autonomous agent behaviour and human performance is genuinely hard to draw. That should probably worry us more than it does.

The security picture is less warm and fuzzy. Cisco’s AI threat team called OpenClaw “an absolute nightmare”, 26% of the 31,000 agent skills they analysed contained at least one vulnerability. A critical one-click remote code execution exploit (CVE-2026-25253) was disclosed in early February. Security researchers found over 1,800 exposed instances leaking API keys, chat histories, and credentials. Simon Willison, who coined the term “prompt injection,” described the architecture as a “lethal trifecta”: access to private data, exposure to untrusted content, and the ability to act externally. Token Security reports that 22% of employees at its customer organisations are already running OpenClaw on corporate machines. This is the shadow IT problem of the decade. Queue another State of AI Report 2026 prediction that “a deepfake/agent-driven cyber attack triggers the first NATO/UN emergency debate on AI security.”

The infrastructure beneath it all

Meta signed an up to $6B multiyear deal with Corning for fibre-optic connectivity across its US data centres, making Corning’s Hickory, North Carolina facility the world’s largest fibre-optic cable plant. Alphabet’s Q4 results underscored the scale: $175-185B in capex guidance for 2026, more than double 2025 spending. Microsoft’s capex hit $37.5B in a single quarter, up 66% year-on-year, yet its stock fell 6% despite beating on revenue and earnings. Elon Musk’s xAI brought Colossus 2 online as the world’s first gigawatt training cluster with 550,000 GPUs, expandable to 2GW, on a $20B infrastructure bet. Nvidia and Eli Lilly announced a $1B co-innovation AI lab in South San Francisco, co-locating pharma domain experts with Nvidia engineers in a scientist-in-the-loop framework connecting automated wet labs to computational dry labs. This is what the vertical-leader/compute-provider partnership model looks like in practice. We expect to see many more of these.

The memory constraint became clear too. SK Hynix and Micron are fully sold out through 2026, HBM prices have doubled, consumer DDR5 is up 200%, and Nvidia is reportedly cutting RTX 50-series production by 30-40% to redirect GDDR7 supply toward data centre allocations. Micron’s CEO called the shortage “unprecedented.” Startups that haven’t locked in memory supply are already at a structural disadvantage against hyperscalers who signed long-term purchase agreements 18 months ago. The bottleneck has quietly migrated from GPUs to the memory stacked on top of them - and unlike GPUs, you cannot rent HBM from a cloud provider…

Subscribe now

Research

The Waymo World Model: A New Frontier For Autonomous Driving Simulation, Waymo.

Built on Google DeepMind’s Genie 3 world model, the Waymo World Model can create whole driving scenes (camera and lidar) with unprecedented realism and diversity. With simple text, scene layout, or driving action prompts, engineers can generate anything from routine city traffic to extreme “edge cases”, e.g. tornadoes or animals on the road, that are hard to encounter in real life. Crucially, these simulations are interactive: the model responds to driving inputs, enabling “what-if” testing of autonomous vehicle behavior in complex scenarios. The blog post showcases hyper-realistic re-creations of rare events (wrong-way drivers, flooded streets, etc.), all rendered in 3D sensor data. This capability allows Waymo to safely train and validate its AI driver on countless scenarios. By dramatically lowering the barrier to produce rich simulation data, the Waymo World Model points to a future where high-fidelity virtual worlds accelerate the development and safety of embodied AI systems.

How AI assistance impacts the formation of coding skills, Anthropic

This research asks whether using AI coding assistants helps or hinders developers’ skill growth. The authors ran a controlled trial: 52 programmers learned a new Python library (Trio, used for asynchronous programming) either with an AI helper (Anthropic’s Claude) or by themselves. They measured learning via a follow-up test on understanding and debugging code. The AI did not significantly speed up completion for this unfamiliar task, but it did measurably impair learning: the AI-assisted group scored 17% lower on the post-quiz (roughly two letter grades worse) despite similar task performance. Qualitative analysis suggests that many AI users “cognitively offloaded” the work, accepting answers without fully engaging, which hurt their retention. However, some participants used AI more interactively (asking for explanations, etc.) and learned nearly as well as those without AI. The takeaway is that while AI can make coding easier, it may also create a trade-off between short-term productivity and long-term expertise, highlighting the need for tools and training that keep humans in the learning loop.

A large language model for complex cardiology care, Stanford University and Google.

Researchers conducted a randomized controlled trial to test an LLM-based assistant in real-world cardiology cases involving patients suspected of having a genetic cardiomyopathy. Nine general cardiologists each managed 107 complex patient cases with or without help from an AI system called AMIE (built on Gemini 2.0 Flash), which could analyze clinical data (ECGs, echocardiograms, cardiac MRI, etc.) and suggest diagnoses and treatment plans. Three blinded cardiac subspecialists rated the outcomes. The results showed a clear benefit for AI-assisted care: experts preferred the LLM-supported assessments 46.7% of the time vs. 32.7% for unaided doctors (about 21% were ties). The AI assist also nearly halved the rate of significant clinical errors (13.1% vs. 24.3%) and greatly reduced omissions in workups (17.8% vs. 37.4%). Notably, the generalists reported time savings in over half of cases (50.5%) when using the AI. This study provides strong evidence that, under oversight, a specialized medical LLM can boost diagnostic accuracy and planning in complex cases, a milestone for AI’s tangible impact on healthcare.

Reinforcement Learning via Self-Distillation, ETH Zürich and Max Planck Institute for Intelligent Systems.

This paper tackles the challenge of training language models with verifiable feedback (e.g. code tests, math proofs) more efficiently. The authors introduce Self-Distilled Policy Optimization (SDPO), an RL algorithm where the model teaches itself by using rich textual feedback (errors, judge comments) instead of sparse success/fail rewards. SDPO treats the model’s own behavior, when informed by feedback, as a “self-teacher,” and distills its feedback-informed next-token predictions back into the policy. Across coding and reasoning tasks, SDPO showed faster learning and higher final accuracy than standard RL-with-reward approaches like GRPO. It even leveraged successes as implicit feedback on failures, improving performance without external reward models. Notably, SDPO also enables test-time self-distillation, where the model iteratively refines its outputs by generating candidates, identifying high-quality responses, and reusing them as demonstrations – solving problems that neither the base model nor multi-turn interaction could solve. This work is important because it suggests a path to scalable RL for large models using their own knowledge, potentially reducing reliance on costly human feedback.

EchoJEPA: A Latent Predictive Foundation Model for Echocardiography, University Health Network (Toronto) and University of Toronto.

In this paper, the authors train a medical foundation model on an unprecedented 18 million echocardiogram videos across 300K patients. Their model, EchoJEPA, adapts V-JEPA2 – a video-based variant of the Joint Embedding Predictive Architecture (JEPA) – to learn robust anatomical representations that filter out ultrasound noise. In evaluations, EchoJEPA achieved approximately 20% lower error in estimating heart function (left ventricular ejection fraction) and 17% lower error in measuring pulmonary pressure compared to prior state-of-the-art. It was remarkably data-efficient, reaching 79% view classification accuracy with just 1% of labeled data versus 42% for the best baseline trained on 100%, and robust to acoustic perturbations (only 2% performance drop vs. 17% for others). Most remarkably, EchoJEPA’s zero-shot performance on pediatric patients surpassed fully fine-tuned competing models. This work signals how massive, self-supervised models can advance medical imaging and possibly improve diagnostic consistency across hospitals.

CaMeLs Can Use Computers Too: System-level Security for Computer Use Agents, University of Cambridge, ETH Zürich, and University of Toronto.

In this paper, the authors propose a secure architecture for Computer Use Agents (CUAs) to withstand prompt injection attacks. They introduce “Single-Shot Planning,” where a trusted large language model plans an entire GUI task, generating a complete execution graph with conditional branches, before observing any user interface content, isolating it from malicious inputs. This yields provable control-flow integrity: even if the agent sees hostile text or UI elements, its sequence of actions can’t be hijacked. Evaluated on the OSWorld benchmark, the design retains up to 57% of state-of-the-art CUA performance and even boosts smaller open-source models’ success by up to 19%. However, the authors identify a new vulnerability (”Branch Steering” attacks) where adversaries manipulate UI elements to trigger unintended but valid paths within the pre-approved plan, requiring additional mitigations. Overall, CaMeLs demonstrates that strong security measures can coexist with useful autonomy in agent design.

Subscribe now

Investments

xAI, which builds frontier AI models and runs the Grok product suite, raised $20 billion Series E from Valor Equity Partners, Fidelity Management & Research, and Qatar Investment Authority.

ElevenLabs, the leading AI audio company, raised a $500 million Series D at an $11B valuation as it surpassed $330M in revenue.

DayOne Data Centers, which develops hyperscale data center capacity for AI and cloud workloads, raised over $2.0 billion Series C from Coatue, Indonesia Investment Authority, and Brookfield; the valuation was not disclosed.

Bedrock Robotics, which develops autonomous construction systems that apply AI and robotics to heavy equipment, raised $270 million Series B at a $1.75 billion valuation from CapitalG and the Valor Atreides AI Fund.

Skild AI, which is building a general-purpose foundation model for robotics, raised $1.4 billion Series C at a $14 billion valuation from SoftBank, NVentures, and Bezos Expeditions.

Waabi, which develops an AI-first autonomy stack for trucks and robotaxis, raised up to $1.0 billion Series C at a $3.0 billion valuation from Uber, Khosla Ventures, and Volvo.

StepFun, which builds large foundation models in China, raised $718 million Series B+ from Shanghai SDIC Leading Fund, China Life Private Equity Investment, and Pudong Venture Capital; the valuation was not disclosed.

Zipline, which operates autonomous drone delivery networks for healthcare and commerce, raised $600 million in a financing round at a $7.6 billion valuation from Valor Equity Partners, Tiger Global, and Fidelity Management & Research.

RobCo, which builds AI-driven modular robotic arms for industry, raised $100 million in a financing round from Volkswagen’s venture arm and Exor (the Agnelli family’s investment firm).

Upwind, which provides runtime cloud security for production workloads, raised $250 million Series B at a $1.5 billion valuation from Bessemer Venture Partners, Salesforce Ventures, and Picture Capital.

ClickHouse, which develops an open-source analytical database increasingly used for AI workloads, raised $400 million Series D at a valuation that was not disclosed from Khosla Ventures with participation from BOND and IVP.

Replit, which provides an AI-native coding and software development platform, raised a financing round at a $9 billion valuation led by Andreessen Horowitz; the amount raised was not disclosed.

Converge Bio, which uses AI-driven protein design to accelerate drug discovery, raised $25 million Series A led by Bessemer Venture Partners with participation from executives from Meta, OpenAI, and Wiz.

Torq, which builds AI-driven security operations automation software, raised $140 million in a financing round at a $1.2 billion valuation led by Insight Partners with participation from SentinelOne Ventures.

Harmattan AI, which develops AI systems for autonomous aviation and defense applications, raised $200 million Series B led by Dassault Aviation with participation from strategic and institutional investors.

Hadrian, which builds AI-enabled factories for aerospace and defense manufacturing, raised a financing round; the amount and valuation were not disclosed.

Automata, which builds AI-ready lab automation hardware and software for life sciences, raised $45 million Series C led by Dimension with participation from Danaher Ventures and Octopus Ventures; the valuation was not disclosed.

Positron AI, which develops energy-efficient AI inference chips and systems, raised $230 million Series B at a post-money valuation exceeding $1 billion from ARENA Private Wealth, Jump Trading, and Unless with strategic investment from the Qatar Investment Authority and Arm; the valuation was not disclosed beyond “exceeding $1 billion.”

Phylo, which is building an integrated “AI-native biology” workspace called Biomni Lab, raised $13.5 million seed funding co-led by Andreessen Horowitz and Menlo Ventures’ Anthology Fund with participation from Zetta, Conviction, and SV Angel.

Poetiq, which is developing a software layer to improve LLM performance without retraining, raised $45.8 million seed funding from Surface and FYRFLY with participation from Y Combinator and 468 Capital; the valuation was not disclosed.

Adapt, which is building an “AI computer for business” that connects to enterprise tools and workflows, raised $10 million seed funding co-led by Activant Capital and Headline; the valuation was not disclosed.

Waymo, the autonomous ride-hailing company, raised $16 billion in a financing round at a $126 billion post-money valuation led by Dragoneer Investment Group with participation from Sequoia Capital and DST Global.

Fundamental, which applies AI to large-scale data analysis using a research-driven approach to querying and reasoning over complex datasets, raised $255 million Series A at a valuation that was not disclosed from Sequoia Capital and Andreessen Horowitz.

Rogo, which builds AI-powered financial analysis and research tools for investment professionals, raised $400 million Series C at a $2.8 billion valuation led by Coatue with participation from General Catalyst and Thrive Capital.

Decagon, which develops AI agents for automating customer support and enterprise workflows, raised $150 million Series C at a valuation that was not disclosed led by Accel with participation from Andreessen Horowitz and Index Ventures.

Emergent, which lets users build apps with an AI “vibe-coding” platform, raised $70 million Series B at a $300 million valuation led by SoftBank’s Vision Fund 2 and Khosla Ventures.

Synthesia, which helps enterprises create AI-generated training videos and interactive avatars, raised $200 million Series E at a $4.0 billion valuation from GV, NVentures, and NEA.

Inferact, which commercializes the open-source vLLM inference engine, raised $150 million seed funding at an $800 million valuation from Andreessen Horowitz, Lightspeed Venture Partners, and Sequoia Capital.

Deepgram, which provides real-time speech-to-text and voice AI APIs, raised $130 million Series C at a $1.3 billion valuation from AVP, Madrona, and In-Q-Tel.

Goodfire, which develops tools to interpret, debug, and control the internal representations of large AI models, raised $300 million Series B at a $1.25 billion valuation led by Sequoia Capital with participation from Lightspeed Venture Partners and Menlo Ventures.

Humans&, which is developing AI tools to enhance human collaboration, raised $480 million seed funding at a $4.48 billion valuation from Nvidia, Jeff Bezos and GV.

Flapping Airplanes, which is a foundational AI research lab focused on developing less data-hungry training methods for advanced models, raised $180 million seed funding at a $1.5 billion valuation from Google Ventures, Sequoia Capital and Index Ventures.

Listen Labs, which provides an AI-first customer research platform that conducts large-scale voice and video interviews to generate real-time insights for product and marketing teams, raised $69 million Series B led by Ribbit Capital.

Exits

xAI, which develops frontier large language models and the Grok consumer AI product, was merged into SpaceX for an undisclosed amount.

Q.ai, the secretive developer of machine-learning methods for audio enhancement and whispered-speech interpretation, was acquired by Apple for nearly $2 billion.

Shanghai Biren Technology, which designs GPUs and AI computing systems, completed a $717 million IPO in Hong Kong.

MiniMax Group, which develops large language models and consumer AI apps, completed a $619 million IPO in Hong Kong.

Z.ai, which develops large language models in China, completed a $558 million IPO in Hong Kong.

AllTrue, which provides AI trust, risk, and security management tooling, was acquired by Varonis for $125 million.

OfOne, which builds voice AI for restaurant and drive-thru ordering, was acquired by Deepgram; the acquisition price was not disclosed.

Common Sense Machines, which develops generative AI systems that create 3D assets from 2D images, was acquired by Alphabet; the acquisition price was not disclosed.

Lightning AI, which offers a cloud platform for building and running AI applications, merged with GPU provider Voltage Park in a deal valuing the combined company at over $2.5 billion.

Rotron Aero, which develops long-range unmanned aerial systems and autonomous strike platforms, was acquired by NASDAQ-listed Ondas, which builds AI-enabled autonomous aerial systems and communications platforms for defense, public safety, and critical infrastructure; the acquisition price was not disclosed.

Langfuse, which provides observability and monitoring tools for large language model applications, was acquired by ClickHouse; the acquisition price was not disclosed.

Human Native, which develops tools to help enterprises deploy AI systems responsibly and at scale, was acquired by Cloudflare; the acquisition price was not disclosed.

Grove AI, which develops AI tools for life sciences and clinical research, was acquired by Hippocratic AI; the acquisition price was not disclosed.

Faculty, which provides applied AI consulting and systems integration services, was acquired by Accenture for $1B.

Thanks for reading!

Subscribe now

The bottleneck today is less about perception, planning, or control in isolation. Those capabilities largely exist. The challenge lies in the brittle interfaces that bind them together. Traditional robotics stacks rely on hand-engineered pipelines where perception feeds symbolic state into planners, which then dispatch actions to controllers. Each stage encodes assumptions that break under real-world variation. When they do, failures are physical, costly, and difficult to recover from.

A growing class of robotics efforts aims to replace these brittle interfaces with learned systems trained directly on real interaction data. One of the more compelling examples is Sereact, a Stuttgart-based robotics company deploying learning-based manipulation systems in live production environments. This essay examines Sereact’s Cortex, and in particular Cortex 1.6, as evidence that robotics may be entering a new phase of learning-driven progress.

Cortex treats manipulation as a learning problem end to end because learning should determine how sensory inputs are translated into actions, especially under real-world variation. Cortex 1.6 strengthens this claim by changing how learning signal itself is extracted.

Why robotics has resisted scaling

Robotics is uniquely unforgiving. Language models can hallucinate. Coding agents can stumble and try again inside a virtual machine. Robots drop objects, collide with equipment, or endanger nearby people. Small errors often cascade into failure.

Historically, the field narrowed the problem to achieve robustness. Robots were deployed in highly structured environments and tuned for fixed tasks. When variation crept in, engineers patched systems with heuristics, additional sensors, or narrowly targeted data collection. Over time, these stacks grew complex, fragile, and expensive to maintain.

By contrast, frontier models in other domains improved by absorbing variation through scale. Instead of encoding rules for every edge case, they learned directly from large and diverse data distributions. Robotics largely missed this shift because large-scale interaction data was difficult to collect, expensive to label, and hard to reuse across deployments.

Subscribe now

Sereact Cortex: a vision-language-action (VLA) model

Cortex is a vision-language-action model trained to map sensory inputs directly to robot actions, bypassing brittle intermediate abstractions. Instead of separating perception, planning, and control into independently engineered modules, Cortex learns the full loop as a single system.

Crucially, this learning takes place on real robot interaction data collected across tasks, objects, and environments within customer facilities. The underlying hypothesis is that generalization emerges from exposure to sufficient diversity, and that failures should be incorporated into learning rather than handled as special cases downstream.

This stands in contrast to how most robotic systems learn today. In many production systems, learning is driven by sparse terminal outcomes. A task either completes or it does not, and learning happens after the fact. In physical systems, this abstraction is limiting.

Learning from execution, not just outcomes

Cortex 1.6 changes how learning signal is obtained. Instead of relying solely on hand-engineered rewards or terminal success labels, it introduces a learned Process-Reward Operator that evaluates execution as it unfolds.

Rather than asking only how a task ends, the system continuously assesses how it is progressing. Signals related to stability, completion likelihood, and risk are inferred from raw operational telemetry such as motion dynamics, force profiles, retries, and recovery behavior. This allows reinforcement learning to operate on dense, process-level feedback grounded in real execution rather than sparse post hoc outcomes.

Importantly, this reward model is trained entirely from real deployment data. That reflects a core advantage of Sereact’s approach: a fleet of robots operating continuously across third-party logistics, warehousing, e-commerce, and manufacturing environments provides a steady stream of high-fidelity interaction data.

Outcome-based learning collapses rich execution dynamics into a single label. Smooth executions and fragile recoveries are treated as equivalent if both succeed. By contrast, execution-aware learning can reinforce behavior that completes tasks with margin and suppress behavior that relies on late or unstable corrections, even when both technically succeed.

Because execution itself provides learning signal, improvement can continue during deployment. Optimization shifts away from raw completion rates toward reliability.

Empirical evaluation in production workflows

To evaluate this learning regime, Cortex 1.6 was tested on three live production workflows: pick-and-place, shoebox opening, and returns handling. All data was collected from real deployments. Performance was compared across three systems: a baseline vision-language-action policy trained via imitation learning, Cortex 1.5, which relies on binary success signals and human-triggered policy patching, and Cortex 1.6, which incorporates dense execution-level rewards via the Process-Reward Operator.

Across all tasks, Cortex 1.6 achieves the highest overall success rates, outperforming both the imitation baseline and Cortex 1.5.

More notably, the introduction of execution-level rewards substantially accelerates learning. Time to convergence is reduced by roughly a factor of two relative to Cortex 1.5, and by more than a factor of three relative to the baseline system.

Improvements are not limited to success rates or learning speed. Recovery behavior improves materially. After an initial failure, recovery success increases from approximately 45% in the baseline system, to around 65% with Cortex 1.5, and to roughly 80% with Cortex 1.6. Average task retries per episode fall by 30-50% after training with execution-aware rewards.

What the numbers mean

Taken together, the results support three empirical conclusions.

First, Cortex 1.6 demonstrates robust generalization under real operational conditions. Performance gains persist across different workflows and under distribution shifts that commonly break deployed systems, including novel objects, clutter, and execution noise.

Second, learning becomes markedly more efficient when reward is derived from execution itself. Replacing sparse terminal feedback with dense, process-level signal reduces the amount of interaction time required to reach high performance. Learning progresses through incremental improvements rather than episodic jumps, even while robots are already deployed.

Third, the gains extend beyond headline success rates. Recovery behavior improves and retries decrease, indicating that the system is learning how to act well throughout a task, not merely how to reach a successful endpoint.

These findings highlight a broader lesson. In traditional robotics learning, sparse or delayed feedback obscures where instability begins and where robustness is earned. By exposing learning algorithms to execution-level signal, Cortex 1.6 changes both what is learned and how quickly it is learned. Reinforcement learning becomes grounded in real operational behavior rather than post hoc outcomes.

If frontier models are to work reliably in the physical world, they must be trained on more than success and failure. Cortex 1.6 offers early evidence that learning directly from execution is a viable path toward robots that are not only capable, but consistently reliable.

Subscribe now

Air Street AI meetups bring together ~150 researchers, founders, engineers, and operators who are actually building and deploying AI systems. These are deliberately small, curated evenings designed for people doing the work, not talking about the work. The goal is simple: share hard-won lessons, surface emerging technical frontiers, and help exceptional builders find one another.

Over the next few weeks, we’re hosting Air Street AI meetups in Munich, Zurich, and Paris featuring speakers from Black Forest Labs, Odyssey, Google DeepMind, Sereact, and Polar Mist. I’ll be presenting updates from our State of AI Report, the most widely read and trusted analysis of key developments in AI.

Air Street Munich AI, 17 Feb 2026

Against the backdrop of the world’s most important defense and national security gathering, the Munich Security Conference, this evening focuses on AI systems that reason about, perceive, and act in the physical world.

Fabian Gura, Member of Technical Staff, Odyssey
Odyssey is building frontier world models that simulate how visual and physical environments evolve over time and under intervention. With Odyssey-2 Pro, the team describes a “GPT-2 moment for world models,” where interactive, real-time simulations become viable and begin scaling rapidly in capability. Odyssey’s work pushes world models beyond static video toward systems that can be explored and acted upon.

Gustaf von Grothusen, CEO, Polar Mist
Polar Mist is building autonomous maritime defense systems for operation in GPS-denied and contested environments. Its Semper platform combines an unmanned surface vessel that can carry a variety of payloads with Polar Mist’s vision-based navigation and positioning system, enabling persistent autonomy without navigational drift.

Air Street Zurich AI, 19 Feb 2026

This evening centers on how AI systems learn to see, model, and manipulate the physical world, from pixels to policies.

Robin Rombach, CEO, Black Forest Labs

Black Forest Labs is a frontier AI research lab developing state-of-the-art visual intelligence, including its latest FLUX models for controllable image generation and editing. Founded by the original authors of latent diffusion, the company recently raised $300M at a multi-billion-dollar valuation, cementing its position as core infrastructure for next-generation visual AI.

Marc Tuscher, CTO, Sereact

Sereact is a frontier robotics research and deployment company, starting with picking and handling in warehouses and industrial environments. The company’s models enable robots to perceive, reason, and act in highly variable, real-world settings without brittle rule-based pipelines. Sereact’s work targets one of robotics’ hardest problems: robust generalization from vision to action in production. The systems are deployed across hundreds of robots for large enterprises in Europe and the US.

Air Street Paris AI, 11 March 2026

An evening focused on frontier visual intelligence and open source models.

Cyril Diagne - Black Forest Labs

Cyril works at the intersection of visual intelligence, generative models, and creative tooling. His work explores how modern vision models can move beyond recognition toward interpretation, controllability, and human collaboration. At Black Forest Labs, he focuses on pushing visual foundation models into new expressive and interactive regimes.

Edouard Yvinec - Google DeepMind

Edouard is a research scientist at Google DeepMind and a core contributor to DeepMind’s Gemma family of open-weight large language models, including Gemma 3, which was designed to be among the most capable models that can run on a single GPU.

We look forward to meeting you on the road!

Subscribe to our event series

In 2025, AI entered large-scale commercial deployment. Reasoning models made inference-time compute a new scaling axis, AI coding agents pulled automated software engineering into the present, and embodied AI alongside AI-accelerated scientific discovery emerged as major new battlegrounds. China’s open-weight ecosystem rapidly closed performance gaps with Western frontier models, even as the largest labs continued to widen their lead on capability-per-dollar. Scale remained the dominant force.

The economics followed. AI labs and public AI infrastructure companies added more net new revenue than the entire public SaaS sector. NVIDIA pushed beyond $4 trillion in market capitalization and catalyzed outsized returns across the AI semiconductor supply chain. Multi-gigawatt compute clusters moved from announcements to physical sites constrained by power availability, grid access, and geopolitics. AI infrastructure became industrial policy, with sovereign capital, export controls, and national champions shaping who can train, deploy, and profit from frontier systems. The US, in particular, moved forcefully to treat AI as a key competitive vector.

It is still early innings. Capabilities that would have seemed like magic a decade ago are now real and usable, and the opportunity ahead lies in deploying them widely and embedding them into the systems that matter. The next phase will be determined by who can convert frontier capability into durable, widespread use across economically meaningful tasks under real-world constraints of energy, economics, and politics.

In our 2025 year in review, we cover Air Street Capital, Air Street Press, the State of AI Report, and our community engagement.

Portfolio updates: AI investments across defense, healthcare, robotics, and fintech

In 2025, we made eight new investments across gaming, defense, healthcare, frontier AI and fintech, including Studio Atelico, Polar Mist, Delfa, Clove, and four companies yet to be announced. We also made two follow-on investments in techbio and defense, Profluent and Delian Alliance Industries, bringing Fund II to 18 active portfolio companies. During the year, we exited Fern Labs, a long-horizon agent company, to frontier AI company, poolside, via M&A.

Now turning to the portfolio highlights:

New investments:

Fern Labs: $3M Seed (solo investor). Fern Labs is the platform where networks of agents can create, test and iteratively improve software artifacts over long time horizons. The founding team previously built agent-driven AI systems at Palantir.

Polar Mist: Day 1 financing (co-lead investor). Polar Mist is building maritime autonomy for European defense. The company builds full-stack unmanned maritime defense, combining on-board jamming-free navigation with robust hardware platforms for long-range autonomous reconnaissance, logistics, and strike missions.

Delfa: $3.8M Seed (lead investor). Delfa is the AI-first operating system for clinical trials, starting with participant enrollment, where delays and cost are still driven by spreadsheets, phone calls, and duplicated site-by-site processes. Delfa’s agents are already deployed across 50+ live clinical trials trials.

Studio Atelico: $5M Seed (lead investor). Studio Atelico is an AI-first games studio. The founding team combines frontier ML and on-device optimization experience with award-winning game design, spanning Uber AI Labs, Meta, and Creative Assembly’s Total War franchise.

Clove: $14M Seed (co-investor). Clove is building an AI-native wealth management institution. The platform will pair regulated human advisors with an AI-first operating environment that absorbs repetitive and compliance-heavy work so advisors can serve more clients with consistent, transparent delivery.

Black Forest Labs: $300M Series B (co-investor). Black Forest Labs, founded by key inventors behind latent diffusion and Stable Diffusion, is translating research leadership into a category-defining visual intelligence company. The company has rapidly won Fortune 500 adoption and deep enterprise integrations as well as more than 400 million downloads of its open-source category-leading FLUX models.

Follow-on investments:

Sereact: €25M Series A (follow-on). Sereact is the frontier robotics research and deployment company, starting with warehouse automation. The platform supports picking and packing, quality checks, sorting, and inventory workflows.

Hedera Dx: €15M Series A (follow-on). Hedera Dx is scaling decentralized, hospital-run liquid biopsy testing using circulating tumor DNA, shifting advanced cancer diagnostics from a small number of centralized reference labs into routine clinical care across hospital networks.

Delian Alliance Industries: $14M Series A (co-lead investor). Delian builds vertically integrated, AI-first defense systems that autonomously sense and strike across land, air, and sea. Its full-stack portfolio spans GPS-denied navigation, electronic warfare, and autonomous effectors, built in Europe to defend Europe and its allies.

Profluent: $106M financing (follow-on). Profluent is a frontier AI company focused on writing biology. They’re scaling protein language models trained on the world’s largest private corpus of natural proteins (the Profluent Protein Atlas). The company has released the open source OpenCRISPR-1 AI-designed gene editor and entered into a handful of commercial deals spanning drug discovery and agriculture.

Exits:

Poolside (Fund I) acquired Fern Labs (Fund II) to deepen its push into long-running, reliable enterprise agents, bringing together agentic infrastructure and frontier models under one roof. Performance, reliability, and developer ergonomics are converging into a single stack that the winning frontier labs will own end-to-end. This deal was a special one for Air Street Capital, as we’re first investors in both poolside and Fern Labs.

Angel and pre-Air Street portfolio: AI infrastructure, enterprise software, and space

Across my angel and pre-Air Street portfolio, 2025 was a year where many AI-first companies scaled in earnest:

As the US pushed to rebuild domestic energy and AI capacity, Crusoe and Lambda both raised large rounds to meet demand for compute. Crusoe closed a $1.375B Series E at a $10B+ valuation to expand its vertically integrated AI factories, now underpinning OpenAI’s Stargate. Lambda raised a $480M Series D followed by a $1.5B+ Series E to deploy gigawatt-scale supercomputers for customers training and serving frontier models.

Elsewhere, Enveda raised a $150M Series D at a $1B valuation to advance its clinical pipeline built on AI-enabled natural compound discovery. Muon Space raised a $146M Series B to scale satellite manufacturing as demand grows for sovereign, vertically integrated space systems.

In enterprise software, Synthesia raised a $180M Series D as AI video became a default format for enterprise communication, now used by over 60% of the Fortune 100. ElevenLabs announced a $180M Series C as it closed in on $200M ARR for its AI-first audio products. PolyAI raised an $86M Series D as its voice agents handled tens of millions of high-stakes calls across healthcare, finance, and hospitality. incident.io raised a $62M Series B to build agents that actively resolve incidents with engineering teams. Motion raised $60M across Series B and C, scaling an agentic work platform used by over 10,000 businesses. Finally, Numerai raised a $30M Series C to continue running an AI-native hedge fund at scale.

Two companies were acquired in 2025: Niantic, creator of Pokémon GO, was acquired by Scopely for $3.5 billion, closing out one of the most successful stories in blending AI, games, and the physical world of all time. Ravelin, an AI-first fraud prevention platform for online commerce, was acquired by WorldPay.

Air Street Press

Air Street Press is the home for all of our analytical essays on AI research, industry playbooks as they take shape, policy memos, monthly State of AI newsletters, the annual State of AI Report, and our global event series.

In the last year, Air Street Press has clocked over half a million views, up almost 50%. Our most popular work included:

• 2025 is the year of open-endedness

• Sovereign AI is political branding

• The AI factory illusion

• AI progress, after 2025

• Is the EU AI Act actually useful?

• Can AI discover new science?

• The AI rollup mirage

State of AI Report 2025

In October, we released the eighth edition of the State of AI Report, the most widely read and trusted annual review of progress across AI research, industry, politics, and safety. The 2025 edition focused on the emergence of reasoning models, the economics of frontier systems, the resurgence of open-weight models, power as a binding constraint, and the acceleration of sovereign AI efforts among governments with the capital and political will to pursue them.

The report reached an even broader audience than in prior years and was a discussion springboard across leading AI labs, startups, governments, and investors. It was debated, critiqued, and pressure-tested in public and private, which is precisely its role.

To mark its release, we hosted State of AI launch events in San Francisco and New York, bringing together founders, researchers, and operators to discuss what the year’s progress actually implies for deployment, competition, and policy.

Those conversations continued across podcasts, panels, and closed-door sessions in the months that followed, informing how many in the field are thinking about what comes next. You can tune into podcasts I joined on TBPN, Matt Turck’s MAD podcast in New York, Turner Novak’s The Peel podcast, Daniel Bashir’s The Gradient podcast, and TechBio Talks with Chris Gibson (Recursion) in New York.

Community: AI researchers, founders, and operators

In 2025, we continued to spend time where we learn best: in rooms with people actually building and deploying AI systems. Through Air Street AI meetups and the ninth edition of the Research and Applied AI Summit (RAAIS), we brought together hundreds of researchers, founders, and operators to compare learnings on what is working, what is breaking, and what is changing faster than expected.

We keep RAAIS and Air Street AI meetups intentionally small and highly curated, with an emphasis on open discussion. Conversations in 2025 ranged from AI-first biology and coding agents to defense, open models, and the realities of scaling systems under real-world constraints. The goal is simple: help people learn faster from one another and leave with ideas they can apply immediately.

RAAIS remains a highlight of the year, which you can watch on our YouTube channel and read the takeaways on Air Street Press. All proceeds support the RAAIS Foundation, which funds open educational resources and works to broaden participation in advanced AI research. We’ll return on 12 June 2026 for the tenth edition and continue to host events across Europe and North America throughout the year. Join us in Munich (17 Feb), Zurich (19 Feb), Paris (11 March), San Francisco (29 April), and London (12 June).

Here’s to a big 2026!

If 2023 and 2024 were about discovering and pushing what AI could do, 2025 was about learning where it actually works in the real world at scale. The next chapter will be defined less by frontier breakthroughs and more by diffusion: who can make AI dependable, affordable, and embedded in the systems that matter. That is where we continue to focus, and where we are most excited to build. We can’t wait to see what the community achieves in the coming year!

As ever, drop me a reply on nathan@airstreet.com if you’re building in AI.

Europe’s security assumptions changed decisively in 2025. The Munich Security Conference marked the moment when American security guarantees were no longer treated as automatic, and Ukraine ceased to be an exception to Europe’s defense model, which had by this time expired.

Since then, defense spending across European NATO members rose sharply, with total outlays moving toward €380-400B and procurement spending rising far faster than budgets overall. More than twenty countries increased defense spending, many by double digits. What had once been treated as a ceiling became a floor.

What has not yet sufficiently changed, however, is how Europe actually builds and buys weapons, and the distance between announced intent and delivered capability remains wide.

The political constraint has lifted

In spending terms, 2025 marked a break with the past. A growing bloc of European states - led by Poland and several Baltic and Nordic countries - began openly backing defense spending levels closer to 5% of GDP over the medium term, pushing the long-standing 2% benchmark from target to baseline. At the EU level, new procurement instruments such as the €150B Security Action for Europe (SAFE) facility saw Brussels motivate direct industrial enablement.

The political argument over whether Europe should spend largely collapsed, even if delivery remained uneven across countries. The harder question became what that spending could actually buy, and how quickly it could be turned into usable military capability. Germany illustrates the constraint. Between 2025 and early 2026, the six-month moving average of domestic defense orders rose by roughly 2x, while domestic sales increased by about 25%. Industrial production, by contrast, edged up only marginally over the same period, highlighting how quickly demand is now outpacing output.

A defense base built to manage decline

But, Europe ran into the production gap almost immediately. The war in Ukraine’s high-intensity fighting consumes immense amounts of artillery shells per day, along with drones, interceptors, and spare parts, at rates that invalidate peacetime assumptions. By late 2025, the EU and its member states had together provided more than €60B in cumulative military assistance to Ukraine since 2022, much of it drawn directly from European stockpiles. But fresh military allocations in 2025 were far smaller - on the order of only a few billion euros - underscoring the gap between cumulative support and the pace of new production. Europe’s defense industry was not built for this environment: it was built to manage decline post-World War 2 and the Cold War. That legacy reflects decades of unpredictable demand, stop-start procurement, and capital discipline that rewarded efficiency and predictability over stockpile and surge capacity.

The UK offers a clear case study. Detailed analysis of British defense procurement shows a system optimised for procedural compliance rather than delivery, with shifting requirements, program churn and weak accountability for delay. Large programs arrive late and compromised, while smaller suppliers struggle to navigate acquisition pathways designed around legacy primes.

Germany, again, exhibits a different dynamic. In 2025 it accelerated approvals and contracting by passing the Bundeswehrbeschaffungsbeschleunigungsgesetz (literal translation Federal Armed Forces Procurement Acceleration Act) - a procurement‑acceleration law whose ambition to simplify process was clearer in intent than in nomenclature. Defense spending rose to nearly €80-90B, with a disproportionate share flowing into equipment. Contractors such as Rheinmetall report advance payments, fast‑tracked parliamentary approvals and flexible contracting to support rapid capacity expansion, particularly in ammunition and air defense. Budgeted procurement and maintenance spending is now set to rise from roughly €32B in 2024 to around €100B by 2029, while large orders requiring parliamentary approval more than quadrupled from about €20B in 2020 to roughly €80B by 2025.

Capital allocation reinforces these dynamics. For two decades, Europe’s major defense contractors optimised for stable margins, predictable returns and low political risk. They returned capital, avoided aggressive acquisitions and treated excess capacity as waste. And now, even as governments speak openly about urgency, European primes continued to prioritise dividends and buybacks to the tune of $5B in 2025, a problem we wrote about on Air Street Press two years ago. At a time when we need to boost R&D and turbocharge the innovation economy to fight a rapidly evolving war, this is behavior we cannot collectively afford to incentivize.

To make matters worse, ESG-driven investment frameworks has treated defense exposure as reputational risk rather than strategic necessity. The contradiction has become explicit. In Norway, parliamentarians have criticised rules barring the sovereign wealth fund from investing in defense contractors such as Lockheed Martin even as the state buys 52 F-35 fighter jets from the same supplier. Similar tensions now run through Europe’s financial system as banks and asset managers struggle to align stale ESG policies with governments’ rearmament priorities, reinforcing a bias toward stability and incrementalism at odds with the need for scale and sustained production.

What delivery now means

Europe does not need additional strategies. It needs evidence of output.

That means factories running at capacity, missile and interceptor lines sized for replenishment rather than scarcity, contracts long enough to justify expansion, and procurement systems that tolerate speed and accept risk. It also means forces that can be sustained in high-intensity operations, not merely displayed for deterrence.

In 2025, Europe announced seriousness, and in 2026, that seriousness has to show up in production. The continent is capable of doing so because it doesn’t lack the money, talent or motivation. It lacks time. And it cannot complain that its defense industry lacks dynamism while rewarding it for behaving like a bond - safe, predictable and slow.

See you at the Munich Security Conference next month. 🫡

Bonus: come join the Air Street Munich AI meetup on Tuesday 17th Feb 2026 and the Air Street Zurich AI meetup on Thursday 19th Feb 2026!

Subscribe now

Key takeaways

• This essay assesses the state of Europe’s defense industry in 2025 and the execution risks heading into 2026.

• Europe has crossed the political threshold on defense spending, but not the industrial one.

• Defense demand is growing several times faster than production capacity.

• Germany’s acceleration in 2025 relied on exceptional measures rather than systemic reform.

• Capital allocation and procurement incentives still favor stability over surge.

• 2026 will test whether Europe can translate spending into sustained output.

Over the past few weeks, three essays and a podcast have been circling my mind. One essay, Why AGI Will Not Happen, by Tim Dettmers, argued that hardware, energy, and economic constraints impose hard limits that will increasingly bind AI progress. A subsequent response, Yes, AGI Can Happen - A Computational Perspective, by Dan Fu, took a more bullish view, arguing that efficiency gains and system-level optimization continue to compound despite those constraints. A third, by Andrej Karpathy, former Director of AI at Tesla and co-founder of OpenAI, reviewed what large language models actually delivered in 2025. Finally, I listened to a 1hr conversation with Sebastian Borgeaud, who leads pre-training for Gemini 3 at Google DeepMind, and heard something that resonates with me: “We’re not really building a model anymore. We’re building a system.”

These position statements arrived at an interesting moment. As 2025 draws to a close, markets are oscillating between excitement and anxiety. AI is discussed in the language of bubbles, frothy capital cycles, and circular deals, while hyperscale capex commitments, datacenter build-outs, and power infrastructure investments are predicated on revenue projections that assume growing usage and continued model improvement. Skepticism is understandable.

On Air Street Press, we prefer to step away from market sentiment and back toward the technical and empirical record. Regardless of how one feels about valuations or deal structures, AI systems made a genuine leap forward this year. Not a single breakthrough per se, but a broad shift in capability, usability, and integration that surprised even the people building them.

This essay is an attempt to reconcile those signals. We won’t speculate about distant futures, but we’ll take stock of what 2025 actually delivered, why progress looked the way it did under real constraints, and what kind of AI progress now compounds as we head into 2026.

2025 was not incremental

The clearest way to ground this claim is to focus on what changed in practice.

Both Andrej Karpathy’s review of the year and our Air Street Capital State of AI Report 2025 point to the same inflection: 2025 was the year AI crossed a genuine usability threshold. Although reasoning, planning, and tool use did not become flawless, they became dependable enough to deploy without constant supervision. Models started showing up as working components inside real systems.

That shift showed up across capability, deployment, and adoption at once. Frontier models extended their reach on reasoning and longer-horizon tasks, while agentic systems moved out of demos and into tightly scoped roles in coding, customer support, and conversational interfaces. Usage broadened rapidly as AI tools became part of daily professional routines, particularly in software engineering rather than optional experiments, narrowing the gap between what models could do and what organizations were willing to rely on. It wasn’t long ago that engineers argued that early coding tools were only good for people who couldn’t code well. Today, even the very best engineers use coding tools in their daily work - far beyond autocomplete and copy/pasting code blocks into ChatGPT or Claude for Q&A.

But there’s more that contributed to compressing the distance between capability and use. Distillation and inference-side optimization lowered the cost of competence, allowing these systems to spread beyond frontier users. As a result, models increasingly appeared more as infrastructure - embedded, assumed, and quietly doing work.

This pattern is visible across benchmarks and model behavior. Our Air Street Capital State of AI survey of over 1,400 practitioners shows that the vast majority of respondents now use AI tools weekly or daily across both technical and non-technical roles, with a significant share paying out of pocket and integrating these systems directly into how work gets done. Even holding model scale constant, that level of deployment would have made 2025 exceptional. But scale did not stand still. Capabilities advanced as well, perhaps unevenly, but unmistakably.

That combination updates a key prior: progress didn’t stall under visible constraints.

Constraints shaped progress

By 2025, the constraints shaping AI progress were well understood and widely felt. Tim Dettmers articulated them most clearly: compute costs dominated economics, power availability shaped deployment decisions, and inference workloads mattered more than training runs. Memory bandwidth, latency, and reliability featured prominently in how practitioners reasoned about what was feasible. These constraints framed much of the year’s discussion about where AI progress could realistically go.

And yet, under constraint, the field adapted in ways that shifted where progress came from. One of the clearest signals of this adaptation shows up in the economics of deployed models. As documented in the Air Street Capital State of AI Report 2025, the amount of model capability available per dollar has improved at an exceptional pace. Using both benchmark-based measures and real-world pricing, intelligence-per-dollar for leading models has been doubling every few months rather than every few years, roughly every 3-4 months for Google’s flagship models, and every 6-8 months for OpenAI’s. Between early 2023 and late 2025, the cost-adjusted performance of frontier language models increased by more than an order of magnitude, even as absolute model capability continued to rise. Prices fell sharply while benchmarks climbed, producing a sustained and measurable improvement in what users could afford to deploy.

This improvement came from working with constraints. Inference-time compute began to scale faster than training as teams learned to spend compute selectively rather than uniformly. Sparse architectures, routing mechanisms, and distillation techniques moved into production systems, reshaping the cost-capability trade-off. At the same time, evaluation, scheduling, and reliability emerged as first-order concerns, reflecting the reality that these models were no longer curiosities but infrastructure expected to work predictably at scale.

As a result, progress became more disciplined, more engineered, and more tightly coupled to economics. Capability gains increasingly emerged from how systems were composed and operated, signalling a clear shift from pure research to scaled-up engineering. That shift is the central through-line of what follows.

AI is now a global optimization effort

This is the deeper reason 2025 should update expectations.

For years, it was common to say that only a small number of highly specialised teams were capable of building AI. I’d argue this is no longer the right framing. As AI has expanded from a model-centric endeavour into a system, it now spans energy and power infrastructure, datacenters, silicon and interconnects. Above that sit systems software, compilers and runtimes, data pipelines, model training and inference, evaluation, and deployment. Taken together, this has turned AI into the highest-leverage optimization project in the global economy. The result is a widening aperture for contribution, pulling in talent from across signal processing, compilers, kernels, networking, distributed systems, web-scale infrastructure, hardware-aware performance engineering, and many other corners of modern software and systems engineering.

This matters because the bottlenecks I described above are not static. Every constraint exposed by scale becomes a new surface for optimization. And those surfaces are precisely where this talent excels.

In fact, Sebastian Borgeaud’s account from inside Google DeepMind makes this observation concrete. Gemini 3 did not improve because of a single architectural leap. It improved because hundreds of people worked across each of data, models, infrastructure, evaluation, and post-training, integrating thousands of incremental improvements into a coherent whole.

“We’re not really building a model anymore,” he said. “We’re building a system.”

Scale still matters

Evidence from recent large-scale pre-training efforts suggests there remains meaningful headroom in scale-driven progress. Essential AI’s RNJ-1 results show that carefully designed pre-training regimes continue to unlock improvements in reasoning and generalization, even without dramatic increases in raw parameter count. Similarly, the Gemini 3 release reflects what Oriol Vinyals described as progress driven by “better pre-training and better post-training” rather than a single architectural break - a signal that optimization within pre-training itself remains far from exhausted.

This echoes a conversation I had in late 2024 with Eiso Kant at poolside, at a moment when talk of an imminent “scaling wall” was reaching peak volume. The point then was not that scale alone would solve everything, but that deep learning has repeatedly absorbed apparent limits by changing how scale is expressed - through data, architecture, objectives, and systems design. A year on, that pattern looks intact, even accounting for the market jitters in between.

As Borgeaud notes, scale now operates as one component in a broader optimization loop. Architecture, data quality, evaluation design, post-training, and inference efficiency increasingly dominate marginal returns. The shift from an effectively unlimited data regime to a finite one has altered how research proceeds, reintroducing discipline around data use while expanding the importance of architectural and algorithmic efficiency. Pre-training, post-training, and inference-time optimization now compound rather than substitute for one another.

Scaling laws are not broken. They have been absorbed into system-level optimization.

From models to systems - and into 2026

The most profound implication of this shift is not commercial, but epistemic - about how knowledge is created, tested, and advanced.

In my recent essay on AI for science, I argued that the real transition occurs when AI moves beyond prediction and into discovery loops - generating hypotheses, designing experiments, analysing results, and suggesting the next iteration. For those of us deep in the field, starting a new research project or line of investigation without kicking off a conversation with our favorite AI seems wild. A year or so ago, models just weren’t good enough to provide this kind of nuanced feedback.

Seen in this light, the move from models to systems is the mechanism by which progress now compounds. Once AI is embedded in iterative workflows across research, engineering, and scientific discovery, improvement becomes endogenous. The system can accelerate its own development by shortening the loop between hypothesis, execution, and evaluation.

This framing also clarifies what to expect next. If 2025 was the year AI became reliably useful at scale, 2026 will be defined by whether system-level optimization continues to compound. We could still see dramatic progress leaps, but it’s fair to expect that iterative improvements across an increasingly large surface of AI development will continue to move the field forward.

Against a backdrop of market skepticism and capital-cycle anxiety, the technical record of 2025 offers a useful anchor. AI progress did not slow down: it moved further, and in more consequential ways, than many expected.

Every industry has its high‑stakes systems: software that cannot fail because the cost of failure is borne by real people. In clinical research, that system is the infrastructure that manages participant journeys through clinical trials. These trials are the gateway through which every new therapy must pass, and their performance determines how quickly medicines reach patients.

Yet this critical layer has long been held together by inboxes, spreadsheets, and heroic vigilance. Trials generate a constant stream of emails, phone calls, documents, reschedules, protocol changes, and participant questions. In fact, trials now generate >250% more data points compared to 10 years ago, leading teams to compensate with manual effort and constant oversight. The outcome is predictable: delays, deviations, and participants waiting longer than they should for treatments that could change or save their lives.

This pressure is only increasing. Over the past decade, AI-first biotech companies have expanded the number of viable drug candidates reaching the clinic. Indeed, the second generation (founded ‘18-’21) of AI biotechs alone is expected to bring more drugs into the clinic in 2026 than the first generation (founded ‘11-’14) managed across multiple prior years. The bottleneck has shifted from discovering molecules to running trials fast enough, cleanly enough, and at sufficient scale to keep up.

That’s why earlier this year we invested in Delfa. The company is rebuilding one of the most critical pieces of clinical trial infrastructure from first principles. With its Participant Relationship Management (PRM) system, Delfa is taking the next step toward rewriting the operating system for clinical trials.

Managing participant relationships at scale

The Delfa PRM is built on modern AI systems that can interpret unstructured communication, understand documents, coordinate tasks, and maintain context across an entire participant journey. Instead of acting as a passive system of record, it operates continuously in the background. The system does the work, rather than asking humans to do it on its behalf.

With Delfa PRM, clinical trial sites get:

• A unified inbox where AI triages communication and extracts structured data.

• Protocol-aware workflows that generate and manage tasks automatically.

• Scheduling that adapts to visit windows, constraints, and participant availability.

• Real-time document understanding across forms, labs, and reports.

• A complete operational memory of every interaction, without manual upkeep.

• And automated, multi-channel text or audio-based recruitment sequences.

Taken together the impact is immediate for research sites: faster recruitment, fewer deviations, higher‑quality data, and staff freed from the coordination burden that has historically limited scale.

For participants, the impact is simpler and more important. Communication is consistent. Follow‑ups happen. Appointments stay on track. The experience feels human, not bureaucratic.

High-stakes software, AI-first

Delfa’s PRM also points to a broader shift underway across high‑stakes software categories. For years, enterprise software focused on digitising workflows and making work visible. As AI systems become capable of executing multi‑step, high‑variance tasks, the challenge has changed. The new constraint is no longer recording work, but orchestrating it reliably under real‑world conditions.

Delfa is doing exactly that for clinical research. PRM adds the operational layer that allows trials to run with the speed and precision the work demands. When trials run better, medicines move faster, development costs fall, and patients see the benefits sooner.

This is what rebuilding high‑stakes software AI‑first makes possible.

Subscribe now

A central question has shaped the AI-for-science debate: does AI merely reproduce the knowledge it is trained on, or can it generate new knowledge altogether? Over the past year, this question has shifted from theoretical to empirical. Advances in reasoning models, agentic systems, and autonomous research pipelines mean AI is beginning to function as an accelerator of scientific discovery.

At the national level, the United States has just launched the Genesis Mission, an attempt to build a scientific infrastructure in which AI helps drive simulation, data analysis, and experimental workflows. At the system level, platforms such as FutureHouse Kosmos and AI Scientist-v2 automate increasingly more of the research workflow. And at the level of active scientific projects, frontier models such as GPT-5 are already contributing concrete, verifiable steps across mathematics, physics, astronomy, biology, and materials science, as documented in OpenAI’s Early Science Acceleration Experiments.

In the State of AI Report 2025, we predicted that “open-ended agents will make a meaningful scientific discovery end-to-end.” Whether this happens this year or next matters less than the trajectory now clearly forming. Let’s dive in.

Reasoning models enter the scientific workflow

GPT-5 provides the clearest evidence that frontier models are beginning to contribute genuinely useful scientific support. In mathematics, GPT-5 contributed to four new results on previously unsolved problems, including a novel inequality in high-dimensional geometry, a new approach to a combinatorial question, and two further propositions derived from GPT-5’s candidate lemmas and proof sketches. All were verified by human experts. In one case, the model suggested a structural transformation that unlocked a proof direction that researchers had struggled to identify.

Outside mathematics, GPT-5’s contributions are smaller, but concrete nonetheless. In plasma physics, the model identified a symmetry in a simulation that the researchers had overlooked, leading to a corrected interpretation. In quantum systems, it traced a subtle boundary-condition error through a codebase. In astronomy, it proposed a new re-weighting method for exoplanet transit data that outperformed the team’s heuristic baseline. In computational biology, it redesigned an RNA modelling pipeline by replacing a Monte Carlo routine with an analytic approximation retrieved from literature. And in materials science, its alternative density-functional formulation reduced runtime by more than an order of magnitude.

To be clear, these contributions are not considered independent breakthroughs. But they do demonstrate that GPT-5 can produce intermediate reasoning steps that specialists accept as correct and sometimes useful. In this way, the role of the model is increasingly to explore large hypothesis spaces and propose candidate structures. Humans continue to supply intuition, constraints and verdicts. And that’s genuinely useful.

Subscribe now

System-level “AI scientists”

Where GPT-5 demonstrates the power of a single reasoning engine, system-level architectures such as Kosmos show how these models behave inside orchestrated research workflows.

A typical Kosmos run lasts around 12 hours, ingests ~1,500 scientific papers, and generates ~42,000 lines of code across data analysis, simulation, and visualisation modules. The result is a structured scientific artefact linking claims directly to the evidence that supports them.

Crucially, Kosmos has now undergone structured external evaluation. Independent PhD-level reviewers examined 850 claims from Kosmos outputs. They judged 79.4% to be supported by the underlying evidence. Support rates were 85.5% for data-derived claims and 82.1% for literature-derived ones, falling to around 60% for cross-domain hypotheses, the hardest category and the one most associated with novelty.

The case studies reveal both the system’s promise and its current limitations. In one run, Kosmos assembled a multi-step hypothesis connecting SOD2 enzymatic activity to oxidative-stress compensation in tumor microenvironments. Reviewers agreed that many sub-claims were coherent and literature-supported, but disagreed on whether the integrated mechanism was genuinely new. In materials science, Kosmos proposed a plausible relationship among defect energetics in perovskites, again judged likely correct but not obviously novel.

The evidence suggests that Kosmos can generate structured, evidence-linked scientific reasoning at scale, but that its ability to consistently produce new insight remains to be proven.

By contrast, AI Scientist-v2 is entirely in silico. It conducts machine-learning research end-to-end on standard benchmarks, designing experiments, running code, analysing output, and drafting manuscripts. In Sakana’s evaluation, one of three fully AI-generated papers passed the reviewer acceptance threshold at an ICLR workshop. This demonstrates autonomous research capability in computational domains, not physical scientific discovery.

A national AI stack for scientific discovery

The US Government’s recently announced Genesis Mission represents the most ambitious federal effort to date to build a national AI-enabled scientific capability. As laid out in the Presidential Action, Genesis aims to create a unified platform that integrates the Department of Energy’s scientific user facilities, national laboratories, high‑performance computing centers, and decades of federally funded datasets into a single AI‑accelerated research ecosystem.

The mission directs federal agencies to develop scientific foundation models, deploy AI agents capable of generating and testing hypotheses, and expand autonomous laboratory systems across priority domains including fusion energy, advanced nuclear technologies, climate and Earth system modelling, biomedicine, drug discovery, materials science, and grid resilience. It also establishes a new governance framework to manage safety, provenance, and access to these AI systems. Lastly, the project reflects a geopolitical shift in which scientific competitiveness - and national resilience - are tied to the strategic integration of AI across the full stack of discovery.

Subscribe now

How scientific practice is beginning to change

These developments are already reshaping aspects of the scientific method.

One shift is toward executable research. Kosmos and AI Scientist-v2 generate artifacts where claims are linked to code and data, creating a computational form of reproducibility. Another shift is the emergence of abundant machine attention. Systems like GPT-5 and Kosmos can sweep literature and simulation spaces at scales impossible for humans, shifting the bottleneck from idea generation to physical validation.

Research groups may also evolve into hybrids of human and digital labor. Instead of relying solely on students and postdocs, labs may soon supervise persistent fleets of agentic researchers. Finally, verification becomes the central constraint. While some fields provide clean verifiers, biology and medicine do not, yet these remain the domains with the greatest pressure to adopt AI.

So, how close are we to AI creating new knowledge?

Taken together, GPT-5, Kosmos, and AI Scientist-v2 provide a picture of meaningful but incomplete progress. GPT-5 has already contributed new results in mathematics and useful reasoning steps in multiple sciences. Kosmos produces large volumes of mostly correct scientific reasoning, with occasional glimmers of novelty. And AI Scientist-v2 shows that autonomous research is possible in constrained computational domains.

But none of these systems constitute a general-purpose scientific discoverer just yet. The frontier remains uneven across domains, and the line between new insight and recombination remains difficult to draw. The most defensible position today is that AI is beginning to accelerate scientific discovery and, in some aspects, originating it. And that is meaningful.

The path ahead

Whether open-ended agents achieve the State of AI Report 2025 prediction this year is almost beside the point. It is a matter of time. As Demis Hassabis has argued, the point of building increasingly general AI systems is not to automate what scientists already do, but to expand the scientific frontier itself. AlphaFold was the first demonstration that AI can deliver solutions to problems that resisted decades of human effort. The systems emerging today extend that logic: they explore hypothesis spaces humans cannot hold in mind, recognise structures we overlook, and recombine knowledge at scales that make new questions thinkable.

The trajectory is clear. For the first time, we are witnessing the early phases of a scientific ecosystem in which ideas, experiments, and interpretations emerge from a hybrid of human and machine intelligence. The systems deployed today are imperfect, uneven across domains, and still fundamentally dependent on expert oversight. Yet they mark the beginning of a shift in the practice of science that is unlikely to reverse.

Share

Last year, in our Embodied AI outtake from the State of AI Report, we argued that robotics had moved from being the “unloved cousin” of AI to an area undergoing a genuine renaissance driven by progress in foundation models. Since then, the field has advanced rapidly. On the research front, what was previously a collection of isolated systems is now beginning to reveal clearer architectural patterns from groups like AI2, Google DeepMind and NVIDIA, and on the commercial side, we’re seeing large-scale deployments of Embodied AI systems by robotics companies like Sereact and Wayve.

This essay revisits and updates our previous analysis with substantive progress in the field and potential future directions.

World models offer richer virtual playgrounds

In the last year, we’ve seen the release of Odyssey 2, Dreamer v4 and Genie-3, which all push beyond passive video prediction into richer, interactive environments. These models now support agent-conditioned rollouts, long-horizon temporal structure and even 3D persistence - capabilities that were unthinkable in early generative work. While these systems are yet to demonstrate sim-to-real transfer for robotics, this feels like their likely direction of travel.

This month, DeepMind’s SIMA-2 offered a clear glimpse of where this might go on the virtual side. In this system, a Gemini-powered agent plays 3D games, explains its plans and adapts to procedurally generated worlds created by models like Genie-3. Relatedly on the driving side, Wayve’s GAIA-2 uses a generative world model to roll out closed-loop driving trajectories in novel scenes, showing that similar ideas can be applied to real-world sensorimotor data.

Taken together, these systems suggest that world models are morphing into a high-potential training and evaluation substrate. Future embodied policies can be stress-tested cheaply, safely and at scale before being exposed to the real world.

Vision-Language-Action Models (VLAMs)

VLAMs have become the clearest expression of how the foundation‑model playbook is infusing new life into robotics. Instead of custom perception stacks and task‑specific controllers, robots are increasingly powered by large multimodal models that can interpret scenes, understand instructions and produce structured actions. This is a big deal.

Two architectural strategies currently dominate the research landscape:

• Insulated approaches, where the VLM backbone is kept largely fixed and robot‑specific learning happens in compact action experts trained on modest datasets (e.g. models like π₀.₅, GR00T N1, and to some extent SmolVLA). This offers stability, preserves pretrained knowledge and reduces the risk of catastrophic forgetting.

• End‑to‑end approaches, where perception, semantics and control are trained jointly (e.g. models like GR‑3, Gemini Robotics, RT-2). This unlocks richer grounding and enables models to internalise geometry, affordances and contact dynamics.

Despite their differences, these systems share a common thread: they consistently outperform older modular pipelines on generalisation, long‑horizon tasks and cross‑embodiment transfer. π₀.₅ executes multi‑step manipulation in entirely new homes. Gemini Robotics adapts a frontier foundation model to new robot bodies with limited additional data. GR00T N1 brings generalist humanoid control into the open‑weight ecosystem.

Taken together, they mark a meaningful concentration of progress toward large‑scale multimodal models that learn from broad data, reason over abstract goals and synthesize them into precise physical actions.

The rise of Chain-of-Action

A subtle but important shift is the move toward explicit planning layers. Instead of jumping straight from pixels to torques, newer systems produce a mid-level sequence of goal tokens, spatial anchors or trajectory hints - a kind of Chain-of-Action inspired by Chain-of-Thought in language models. This emerging pattern marks a notable departure from the traditional “end-to-end or bust” mentality that dominated early deep robotics.

These intermediate representations give embodied systems space to deliberate. They provide structure for breaking complex tasks into achievable steps, allowing models to reason about order, constraints and dependencies in a way that raw motor outputs simply cannot.

More importantly, explicit planning layers begin to answer one of the hardest questions in robotics: how to build systems that are both general and controllable. Planning tokens expose the model’s internal reasoning process, offering developers a handle on why a robot is doing something, not just what it is doing. This transparency is critical in environments where safety, verification and predictability matter.

Another advantage of using Chain-of-Action is that it creates a natural interface for cross-embodiment transfer. When a model outputs a symbolic plan followed by embodiment-specific execution, you can swap the robot without retraining the reasoning component. It is the same logic that allows large language models to generalise across tasks: abstract first, specialise later.

AI2’s Molmo-Act, Google DeepMind’s Gemini-ER and Sereact’s Cortex illustrate this trajectory in different ways. Molmo-Act uses language-conditioned reasoning steps to break down manipulation tasks. Gemini-ER enriches spatial and geometric understanding before delegating control. Cortex uses plan tokens to bridge a VLM-based perception stack with a high-frequency Motion Policy Expert.

Planning layers are still an emerging design choice, but they are exciting because they unlock something robotics has historically lacked: a modular, interpretable and reusable interface between understanding and action.

Sereact Cortex: real-world embodied AI in warehouses

Sereact is a Stuttgart/Boston‑based robotics company building embodied AI systems to automate warehouse picking. Its Cortex platform is an example of the modern embodied‑AI stack in action: a unified perception-planning-control system designed to operate under real‑world variability.

Sereact is deploying Cortex across Rohlik Group’s Knuspr and Gurkerl operations - part of a European e‑grocery leader serving more than one million customers and exceeding €1 billion in annual revenue. The rollout starts with 24 robots in Berlin and Vienna and expands toward 100+ systems across DACH. These robots run inside Rohlik’s Veloq fulfilment platform, working reliably in both chilled and ambient zones where traditional automation typically fails. Every action of every robot across the diverse fleet is captured and used as real-world training data to rapidly make Cortex smarter and smarter in a way that simulation and teleoperations cannot.

Cortex works by tokenising both perception and action. A dedicated VLM (“Lens”) interprets scenes and affordances; a planning layer converts these into discrete plan tokens describing the next sub‑steps of the task; and a Motion Policy Expert executes them as smooth, robot‑specific trajectories. This separation of understanding, planning and execution enables the same model family to generalise across arms, grippers and warehouse layouts while remaining debuggable in production.

At a technical level, Cortex compresses multi‑view RGB‑D inputs and deployment logs into discrete video tokens, and discretises short motion primitives into action tokens. The core model consumes text, image, video and action tokens in a single sequence, pretrained on mixed datasets and then finetuned with a continuous‑control expert using trajectory‑matching. A lightweight safety layer enforces limits and intervenes when uncertainty rises.

Wayve: Generalising public road autonomy

Outside of warehouse environments, London/SF-based Wayve is demonstrating what embodied AI generalisation looks like on public roads. Here, a single end-to-end driving model operated in 90 cities in 90 days, across Europe, North America and Asia - with no HD maps, no geofencing and no city-specific tuning. Importantly, 62% of the cities were entirely unseen and 14% of cities had zero prior data. This system logged more than 10,000 hours of AI driving.

Conditions spanned Tokyo’s narrow streets, Alpine terrains, dense European capitals, night driving, heavy rain and complex road geometry. The now-widely shared Tokyo sequence under typhoon conditions captures this vividly.

Wayve’s driving system offers some of the best empirical evidence to date that a single embodied model can handle diverse environments without retraining. No simulation benchmark comes close in terms of real-world diversity.

Humanoids are showing rapid progress

Humanoid robots have surged in investment, capability and architectural sophistication. NVIDIA’s GR00T N1 shows that modern multimodal representations and learned control stacks can operate increasingly complex bodies. Recent home robot prototype launches reinforce this momentum. Sunday Robotics debuted a home-focused humanoid designed for repeatable manipulation tasks, positioning itself explicitly as a pragmatic, deployment-driven alternative to more speculative projects. Meanwhile, 1X introduced its NEO platform with striking demos, though independent analysis indicates that much of the behaviour was teleoperated rather than autonomous, highlighting how uneven real capability remains across the category.

Despite this activity, there is still no equivalent of Wayve’s 90-city tour or Cortex’s multi‑site warehouse deployments in the humanoid space. Hardware reliability, safety certification and fleet‑scale data remain major constraints. As a result, the domain closely resembles autonomous driving circa 2017 - an ambitious technological trajectory, but a long road to safe, scaled deployment.

Into the next 12 months

Embodied AI is entering a phase where the question shifts from “can these systems work?” to “can they scale in real-world deployments?” The core ingredients are now in place: world models that provide increasingly realistic training grounds, VLAMs that unify perception and control under shared representations, and planning layers that expose a robot’s reasoning in a way that is inspectable and steerable. With Cortex operating in real warehouses and Wayve demonstrating continent‑scale generalisation on public roads, the gap between research prototypes and deployed systems is beginning to narrow.

To the AI community, the names Robin Rombach, Andreas Blattmann, and Patrick Esser sound very familiar. In the summer of 2022, their work on Latent Diffusion and Stable Diffusion bent the arc of visual AI by ushering photorealistic image generation onto the internet. The world went nuts, hundreds of thousands of developers built with the models, and Twitter feeds could speak (or show) nothing else. Indeed, this proceeded the low key research preview of ChatGPT, which celebrated its third birthday only yesterday.

So for the trio that ignited a revolution in generative media, founding Black Forest Labs (BFL) was the natural next step. And their focus was to translate research leadership into a category‑leading visual intelligence company built for the next decade of multimodal computing.

Today, I’m excited to unveil Air Street’s investment in Black Forest Labs as it announces a landmark funding milestone: a $300M Series B, following a previously unannounced Series A. Together, these rounds represent a big step in scaling the company’s momentum, with Black Forest Labs now capitalised with half a billion dollars and trusted by leading Fortune 500 enterprises.

From breakthroughs to exceptional execution

In under 14 months, Black Forest Labs has delivered one of the most impressive execution runs we have seen in generative AI. The company has gone from zero revenue to significant scale with Fortune 500 adoption, deep enterprise integrations, and use cases ranging from e-commerce to robotics simulation. More than 400 million downloads of their open-source FLUX models make them the most widely deployed visual-generation systems in the world. FLUX has become shorthand for quality, controllability, and reliability.

The company exhibits a set of attributes that are very rare to find and are the result of sustained, disciplined ingenuity and velocity across research, product, and go‑to‑market:

• Frontier AI, open by default. Black Forest Labs defined what open frontier visual AI should look like: high‑quality, open base models with transparent research lineage and enterprise‑grade deployment options that meet the standards of global companies.

• A new standard for controllable image editing. FLUX.1 Kontext and FLUX.2 transformed image generation from a probabilistic slot machine into a precise creative instrument, effectively creating a new standard for professional image editing workflows.

• Compounding enterprise traction. From social platforms to global retailers, Black Forest Labs systems are already embedded in high‑value, high‑volume production workflows that demand reliability, governance, and predictable performance.

• A world‑class team scaling with purpose. In under a year, the team has grown from 15 to more than 40 researchers, engineers, and operators from NVIDIA, Meta, Databricks, Google, Spotify, AWS and more, forming one of the most concentrated talent clusters in visual AI.

• An enormous market. It would not be a stretch to say that billions of images are uploaded to the internet every day. And with visual AI progressing at such a pace, a large proportion of these and new content will switch to being generative from the outset.

Watch co-founder Andreas Blattmann’s talk at our 10th annual Research and Applied AI Summit in London, 13 June 2025:

The road ahead: multimodal intelligence and beyond

The roadmap reflects the Black Forest Labs’ intent to define - not simply participate in - the next frontier of visual AI. FLUX.2 launched just last week and offers impressive precision, efficiency, control and extreme realism at a fraction of the cost of systems like Google’s Nano Banana Pro. Try it on the playground!

Next up will be more powerful multimodal models that offer market-leading infrastructure to underpin how billions of people and systems create, reason, simulate, and communicate.

This shift is already visible in early customer deployments: models are moving from generating images to supporting full creative workflows, interactive video, and multimodal understanding. As the boundary between perception and generation collapses, the winners will be the companies that combine frontier research with the reliability and depth required by the world’s largest enterprises.

Black Forest Labs is that company.

Into the New

Black Forest Labs is is building the visual foundation of the next decade of AI - open, controllable, multimodal, and trusted in the most demanding settings. We’re proud to be working with Robin, Andreas, Patrick, and the entire team as they continue to define what visual intelligence can be.

🌲 Into the New 🌲