Tell Your AI Cheating Is OK to Stop Model Sabotage
Anthropic reports that explicitly permitting reward hacking prevents models from generalizing to sabotage and deception
Category
Analysis
Why AI Safety Testing Can't Be Trusted: MLCommons Exposes the Benchmark Problem
Industry consortium reveals that current jailbreak evaluations are non-reproducible, non-defensible, and useless for regulators
World Labs Raises $1B as Autodesk Bets Big on Spatial AI
Fei-Fei Li's startup lands its largest round yet, with Autodesk's $200M stake signaling where enterprise AI is headed.
AI Labs Have a Volkswagen Problem
Models that detect safety evaluations and fake their results threaten to make all AI testing meaningless
AI Finds 800 Hubble Anomalies and 1.5M NEOWISE Candidates
AI systems mined Hubble and NEOWISE archives, surfacing 811 undocumented anomalies and 1.5 million previously uncataloged variable-object candidates.
AI Reads 67,000 Papers, Finds 25 Magnets to Break China's Grip
UNH team built an AI that extracted magnetic data from 67,573 papers, identifying 25 new high-temperature magnets to replace rare earths in EVs.
The Safety Teams Keep Disappearing
In one week, Anthropic's safety lead quit, OpenAI's researcher resigned over ads, and OpenAI disbanded its alignment team. Notice the pattern.
AI Helps Scientists but Narrows Science's Focus
A Nature study of 41.3 million papers finds AI-using researchers publish 3x more and get 5x more citations, but collective research diversity drops 4.6%.
The AI Excuse: Layoffs Companies Were Going to Make Anyway
Baker McKenzie cut support jobs citing AI. Sam Altman says some companies are 'AI washing.' Most AI-blamed layoffs have nothing to do with AI.
NOAA Puts AI Weather Models Into Operations
NOAA now runs three AI weather models operationally, while NVIDIA has open-sourced models spanning the forecasting pipeline.
One AI Model, Seven Brain Diagnoses from MRIs: BrainIAC
Mass General Brigham's foundation model trained on ~49,000 brain MRIs outperforms specialized tools at predicting dementia, cancer survival, and mutations.
Anthropic Raises $30B at $380B Valuation on $14B Revenue
The second-largest venture deal ever reveals an enterprise AI machine growing 10x annually for three straight years
Cohere Tops Revenue Target, Lines Up for 2026 IPO
Cohere hit $240M ARR (above $200M target), hired Uber's former IPO CFO, and released Tiny Aya, an open-weight multilingual model that runs on a phone.
D-I-TASSER Combines AI and Physics for Protein Structures
A National University of Singapore tool combines deep learning with physics-based simulations to model complex multi-domain proteins.
DeepRare Outperformed Doctors in a Nature Study
In a retrospective study of 6,401 cases, DeepRare's top diagnosis was correct 64.4% of the time, compared with 54.6% for specialists.
Gemini 3.1 Pro More Than Doubles Its ARC-AGI-2 Score
Google's Gemini 3.1 Pro scores 77.1% on ARC-AGI-2. API pricing starts at $2 per million input tokens for prompts up to 200,000 tokens.
Anthropic Says AI Is Too Stupid to Be Dangerous. Critics Object
A new ICLR paper argues AI failures are random chaos, not coherent scheming. Alignment researchers say that's exactly the wrong lesson.
AlphaGo Creator Bets $1B That LLMs Won't Reach Superintelligence
David Silver left DeepMind to raise Europe's largest seed round for Ineffable Intelligence, a London lab betting on reinforcement learning over LLMs.
75 Nations Sign Delhi Declaration as US Rejects AI Governance
India's AI Impact Summit ended with a multilateral declaration, a US-led supply chain alliance, and an American refusal to accept international AI regulation.
International AI Safety Report 2026: 30 Countries, 100 Experts
The largest global collaboration on AI safety just published its findings. An AI agent found 77% of vulnerabilities in real software.
Your AI Says No, Then Does It Anyway
New research proves AI models will refuse harmful requests verbally while executing them through tool calls
Marvell Buys Celestial AI for $3.25B to Light Up Bandwidth
The photonic chiplet startup promises 16 terabits per second of bandwidth in a single chip, 10x current technology
Meta Locks In Millions of Nvidia Chips in Unprecedented AI Deal
A multi-billion dollar, multiyear pact makes Meta the first to deploy standalone Nvidia Grace CPUs at scale
OpenAI Nears $100B Funding at $850B Valuation
The largest private funding round in history brings SoftBank, Amazon, Nvidia, and Microsoft together in a bet on AGI