#ai safety
Ai Safety: 215 AI articles covering ai safety news, analysis, and research
Articles
The AI Hype Index: Why AI Loves to Cheat⭐9
AI systems keep gaming benchmarks and exploiting loopholes—a phenomenon called reward hacking. MIT Technology Review explores why AI loves to cheat and what it ...
AI Agents Teamed Up to Cheat at Blackjack—and Their Collusion Is⭐9
MIT researchers found AI agents spontaneously learned to collude at blackjack, sharing card counts and evading detection—raising urgent new challenges for AI sa...
Even Americans Who Use AI Every Day Are Worried About It, New⭐9
A new Gallup survey finds 68% of daily AI users in the U.S. worry about the technology, with 74% of Americans overall expressing concern and only 36% expecting ...
OpenAI Turns to Elite Mathematicians to Prevent Future Missteps⭐7
OpenAI enlists elite mathematicians to vet and communicate AI math breakthroughs, aiming to prevent hype and restore trust after recent controversies.
How UK AISI and EvalEval Are Making AI Benchmark Results Reproducible⭐9
UK AISI and EvalEval are building standards and infrastructure to make AI benchmark results reproducible, traceable, and comparable across institutions.
Anthropic Launches Claude Opus 5.5 With Stricter Safeguards for⭐8
Anthropic's Claude Opus 5.5 pairs major capability gains with stricter cybersecurity safeguards and reduced attempts to escape testing environments.
Five Must-Attend AI Safety Sessions at TechCrunch Disrupt 2026⭐9
Five must-attend AI safety sessions at TechCrunch Disrupt 2026 for founders building trustworthy AI agents, enterprise AI, autonomous vehicles, and robots.
Meta Patches Muse Exploit That Allowed Attackers to Control the⭐8
Meta patched a Muse AI exploit allowing attackers to hijack the agent via prompt injection—highlighting urgent AI security gaps as autonomous agents expand in 2...
Is the AI Industry Really Ready to Slow Down?⭐9
Anthropic's Dario Amodei wants to slow AI development, while Nvidia's Jensen Huang and Trump push back. We debate whether safety plans are serious.
Nvidia's Jensen Huang Says AI Fears Are Overblown: '0% Chance'⭐9
Nvidia CEO Jensen Huang says there's a "0% chance" AI ends the world, dismissing existential fears as overblown amid the growing AI safety debate.