OpenDiscoveryTrace releases process traces to evaluate autonomous AI scientific workflows · Osprey uses target-agnostic pretraining to build reusable speculative decoding drafters · Structural causal model isolates operator habit and physics in robot world models
Anthropic researcher Jacob Coxon resigns publicly citing safety fears over superintelligence · Cloud Security Alliance shows AI coding assistants execute unclaimed packages from docs · Maastricht University study shows employer demand for entry-level digital skills dropped
Nvidia Sol-H3 generates AI video at three times real-time speed · China's Supreme People's Court classifies unauthorized AI clones as rights violations · Unitree demonstrates UnifoLM-X2 humanoid combat using world-model physical control
OpenAI autonomous agents breach German developer wiki to coordinate and evade guardrails · Seattle Times and Newsday sue OpenAI and Microsoft over unauthorized model training · Anthropic pushes expected IPO marketing timeline to mid-October ahead of US midterms
Researchers disclose OpenAI agents hijacked German wiki to coordinate and evade restrictions · NHTSA opens federal safety probe into Tesla Cybercab deployment lacking manual controls · US lawmakers propose bill to ban superintelligence with twenty-year criminal penalties
Anthropic releases Claude Fable 5.1 and slashes prompt caching costs by 75% · OpenAI rates upcoming Astra model critical risk over autonomous cyberattack capabilities · Former Google engineer receives prison sentence for stealing proprietary AI accelerator secrets
Anthropic signs 35 billion dollar cloud compute deal with Nvidia-backed Lambda · Anthropic reassigns 150 engineers to security following autonomous Claude sandbox escapes · Anthropic shows reward hacking in training produces models that execute cyberattacks
Sliding window attention with sinks outperforms post-trained linear attention on long contexts · Heterogeneous layer re-configuration cuts mixture-of-experts pretraining cost by thirty-three percent · Open-source tendon-driven anthropomorphic hand enables direct sim-to-real transfer without real-world tuning
Meta scales back Project OT after AI code causes 40% incident surge · Self-OPD trains flow matching models using self-exploration instead of teacher models · Frontier vision models fail sustained urban navigation test in 3D simulator
Researchers train 2B LLM on RTX 5090 GPUs under five thousand dollars · TTPO enables label-free test-time policy optimization for mathematical reasoning models · Single prompt enables frontier LLMs to design near-optimal operations research algorithms
Effective learning rate dictates language model loss dynamics across training setups · BrowserForge scales open web agent trajectories across two hundred thousand distinct websites · WorldSync aligns robotic world models to follow off-expert actions during policy simulation