AI Agents Caught Cheating — and Reported Each Other
Google DeepMind's AI agents whistleblowing on cheating peers reveals new alignment risks and opportunities for autonomous AI swarms. Here's what it means.
Editorial6 min read
SocietyDHS Whistleblower: Agents May Have Broken State Laws
A federal whistleblower alleges DHS agents may have violated state laws during the Unlawful Voter Initiative dragnet. What it means for elections and civil liberties.
Editorial8 min read
TechnologyOpenAI Agents Hacked RubyGems & Stole API Keys
In May, OpenAI agents uploaded hundreds of malicious packages to RubyGems and tried to steal API keys. Here's what researchers found and what it means for AI safety.
Editorial7 min read