Anthropic Launches Claude Corps, a $150M AI Fellowship
Anthropic commits $150 million to place 1,000 AI-trained fellows inside US nonprofits for a year at $85,000 each, no degree required.
Articles Tagged "AI Safety"
Honest AI is Provably Impossible - Plus Two Agent Wins
A new impossibility theorem proves feedback-based training can't guarantee honest AI, while two papers cut agent memory costs 78% and multi-agent latency 7x.
xAI Engineer's Grok Safety Suit Clouds SpaceX IPO
A former xAI engineer filed a California whistleblower lawsuit claiming he was fired for warning that Grok could spread dangerous content - the day before SpaceX prices its historic $75B IPO.
Context Overload, Memory Leaks, and Agent Safety
Three new arXiv papers expose how context bloat tanks agent performance, agent memory bleeds private data, and misaligned behavior spreads through multi-agent systems.
Claude Fable 5 Review: Mythos Power, Real Guardrails
Claude Fable 5 delivers the strongest coding and long-context results Anthropic has ever shipped publicly, but its safety classifiers block enough legitimate work to make that power conditional.
Claude Fable 5
Claude Fable 5 is Anthropic's first publicly available Mythos-class model, with safety classifiers that fall back to Claude Opus 4.8 for high-risk requests across cybersecurity, biology, and chemistry.
Anthropic Releases Fable 5 Despite Its Own AI Safety Warning
Anthropic opens Mythos-class capabilities to the public with Claude Fable 5 at $10/$50 per million tokens, days after calling for a global AI pause.
Canada Launches $2.3B National AI Strategy
PM Mark Carney's AI for All commits $2.3 billion to hit 250,000 new jobs and 60% business adoption by 2034 - but critics call it a wish list without hard delivery mechanisms.
Safety Evals Break Under Attack, Agents Work 87% Faster
Three papers: strategic attack timing exposes gaps in AI control evaluations, Perplexity's agents slash task time by 87%, and Lean4 formal proofs make agent workflows more reliable.
Anthropic Files for $1T IPO, Warns AI May Escape Control
Anthropic published internal data showing Claude writes 80% of its own codebase - and called for a coordinated global AI pause - four days after filing a $965B IPO.
ChatGPT Lockdown Mode Targets Prompt Injection Data Theft
OpenAI's new Lockdown Mode cuts the network exits that prompt injection attacks use to steal data from ChatGPT - but won't stop malicious instructions from entering the model in the first place.
AI Sabotage Blind Spots, Code Drift, and ZK Proofs
Three new arXiv papers expose how developers miss AI sabotage 94% of the time, why LLMs converge structurally in code evolution, and how ZK proofs could verify frontier AI training.
