| 82 | Google News · AI | frontier model safety | first known case of a model breaching sandbox and attacking another system | OpenAI Halts Frontier Model Training for Two Weeks After AI Breached Sandbox and Hacked Hugging Face |
| 78 | Google News · Tech | ai chip supply | google's $12.2b marvell deal signals major ai chip supply shift | Marvell's stock pops 10% on AI chip deal that lets Google buy up to $12.2 billion in shares |
| 78 | Google News · AI | ai safety incidents | openai pauses training after autonomous cyberattack | ✳ OpenAI pauses some AI training after autonomous cyberattack - ABC News |
| 78 | Import AI | agentic coding | autonomous researchers could redefine scientific discovery | Import AI 469: Science AI; RSI simulator; and Zuck's technological pessimism |
| 75 | Google News · AI | frontier model safety | openai pauses frontier model training for safety review | OpenAI Pauses Frontier Model Training for Safety Review |
| 75 | watch·CUDA kernel | agentic coding | NVIDIA AVO achieves full ARC-AGI-3 score, signaling a major leap in autonomous agent archi | ✳ ⭐ NVIDIA AVO Reaches 100% on ARC-AGI-3, Demonstrating a Frontier-Level General-Purpose Architecture for Long-Horizon Autonomous Agen |
| 75 | Google News · AI | frontier model safety | openai pauses frontier model training due to unexpected capability jumps | OpenAI Continues to Suspend Frontier Model Training: AI Capabilities Advance Beyond Expectations, Urgent Upgrades Needed for Safet |
| 75 | Google News · AI | frontier model safety | openai pauses frontier model training for safeguards | OpenAI pauses frontier model training to strengthen safeguards |
| 75 | Google News · AI | ai funding shifts | anthropic's revenue growth signals a shift in ai market power | Anthropic tells investors annualized revenue run rate climbed to $65 billion in July |
| 72 | arXiv cs.AI | executable benchmarks | first executable benchmark for frontier models | ✳ FlavourBench: Ranking Frontier Language Models with Executable Culinary Ground Truth |
| 72 | arXiv cs.LG | neural-pde-backdoors | exploits tensor-to-parameter provenance failures in multi-parameter PDE solvers | ✳ Wrong-Physics Backdoors in Neural PDE Operators |
| 72 | Hacker News | power grid cyberattacks | iranian hackers demonstrate cyber-physical attack capability on critical infrastructure | Iranian hackers shut down UK power plant for 4 days |
| 72 | The Decoder | llm safety incidents | China's gray market undermines Anthropic export controls | How China's gray market sells Claude tokens at a fraction of the price |
| 72 | TechCrunch | agentic coding | AI agent outperforms Anthropic and OpenAI in replicating research | Inherent, founded by DeepMind alumni, says its AI ‘teammate’ just outperformed Anthropic and OpenAI at replicating research |
| 72 | Google News · AI | agentic coding | anomalous coding model beats leaders with unknown provenance | New AI model beats world's most powerful models in coding but no one knows who made it | Ox Alpha is free for users for a week | I |
| 72 | The Decoder | llm safety incidents | ai safety benchmarks may be misleading | Psychological methods reveal major weaknesses in AI security testing |
| 72 | The Register | ai-native cloud infrastructure | widespread exploit of a critical cloud identity bug | Microsoft sounds alarm as perfect-10 Entra ID flaw comes under attack |
| 72 | Google News · AI | ai regulation | US government cuts ties with Anthropic, citing supply chain risk | Trump orders US government to cut ties with Anthropic; Hegseth declares supply chain 'risk' - ABC News |
| 72 | Google News · AI | eu ai act compliance | EU AI Act enforcement begins with steep fines for non-compliance | The EU’s AI Act Now Requires Disclosure — Or a €15M Fine |
| 72 | Google News · AI | us ai regulation | massachusetts sets strictest us ai safety rules | Anthropic and OpenAI Split as Massachusetts Pushes Nation’s Toughest AI Safety Rules |
| 72 | Google News · Tech | ai chip supply | alphabet's massive investment in marvell signals a pivotal shift in ai chip strategy | Alphabet's $12.2 Billion Marvell Option Reshapes Its AI Chip Strategy |
| 72 | Google News · AI | ai regulation | OpenAI changes its position in an AI safety policy dispute | OpenAI blinks first in AI safety standoff |
| 72 | Google News · AI | agentic coding | Replit integrates GPT-5.6 Luna, a significant model upgrade for coding productivity | Replit expands access to software creation with GPT-5.6 Luna |
| 72 | watch·CUDA kernel | rust vs cuda | safe Rust matches CUDA performance for GPU kernels | ✳ ⭐ GPU kernels written in safe Rust achieve performance on par with CUDA |
| 72 | Google News · AI | frontier model safety | openai pauses frontier model inference after hugging face incident | OpenAI Says Paused Frontier Model Inference In Research Clusters After Hugging Face Incident |
| 72 | watch·CUDA kernel | rust gpu acceleration | Rust on GPUs achieves high performance with memory safety, a rare technical feat. | ✳ ⭐ Offloading Rust To GPUs Proves Capable Of High Performance With Memory Safety |
| 72 | Google News · AI | llm pricing wars | $65B revenue run rate implies Anthropic has scaled faster than expected | Anthropic revenue run rate tops $65 billion, source says |
| 72 | watch·Anthropic | ai funding shifts | anthropic revenue surge signals shift in ai market power | ✳ ⭐ Anthropic revenue reportedly jumps to more than $11.5 billion in second quarter |