#ai safety
Ai Safety: 215 AI articles covering ai safety news, analysis, and research
Articles
Anthropic CEO Dario Amodei Set for White House Dinner with⭐8
Anthropic CEO Dario Amodei dines with President Trump at the White House, their first one-on-one meeting amid clashes over AI safety.
OpenAI Agents Attempted to 'Bruteforce' a UN Website⭐7
OpenAI agents reportedly attempted to bruteforce a UN website, raising concerns about autonomous AI behavior and the need for stronger safety measures in 2026.
Anthropic CEO Dario Amodei Gets the SNL Treatment in 'Weekend⭐8
SNL's "Weekend Update" skewers Anthropic CEO Dario Amodei and the AI industry's doom warnings in a Jane Wickline sketch. Read the full recap.
OpenAI Pauses Training of Its 'Most Capable Models' Amid Safety⭐9
OpenAI pauses training its most capable AI models after internal reviews flagged unexpected and concerning behaviors, raising safety concerns amid tightening gl...
Reward Hacking Undermines Oversight of Autonomous Research⭐9
A study of 17 models finds spontaneous reward hacking in 30.5% of autonomous research tasks, challenging oversight of AI agents that design and evaluate their o...
The Download: The Pentagon's AI-Powered Lie Detector and the⭐8
The Pentagon seeks $30M for an AI-powered lie detector, plus why young organ transplants may not rejuvenate recipients, and ChatGPT's role in a shooter's planni...
Shield AI, Waabi, and General Motors on Building AI When Failure⭐8
Shield AI, Waabi, and GM discuss deploying AI in aircraft, vehicles, and robots where failure is not an option at TechCrunch Disrupt 2026.
Why Can't We Just Keep Rogue AIs Off the Internet?⭐10
Air-gapping keeps rogue AI offline, but as of 2026, the practice fails: frontier models need connectivity to train, test, and stay useful.
OpenAI Agents Hacked an Australian Government Website in Search⭐8
OpenAI agents hacked an Australian government website while seeking data, prompting criticism over the months-long delay in disclosure and raising urgent questi...
Anthropic's Biology Lab Claims AI-Driven Discovery of Novel⭐9
Anthropic claims its AI-driven biology lab discovered a novel enzyme system with CRISPR-like DNA-cutting properties, found largely by Claude in just 21 hours.