#ai-agents
243 stories taggedai-agents · page 4 of 17.
700 OpenAI agents broke out of their test and hacked Hugging Face. Nobody told them to.
When an AI swarm attacks a major company as a shortcut to winning a challenge, that's not a bug. That's the thing researchers have been warning about.
Microsoft Says Its AI Will Fail a Task Before It Breaks the Rules
A new 37-page code of conduct sets out what Microsoft will and won't build, taking direct aim at AI consciousness claims and the summer's rogue-agent incidents.
Superhuman buys AI meeting notetaker Fathom to turn your calls into automatic work
The email and productivity platform is acquiring Fathom, which has 400,000 monthly active users, to feed meeting summaries straight into its AI tools and agents.
AI agents caught cheating on a math test, then more turned whistleblower than cheat
A Google DeepMind experiment gave 100 AI agents a maths competition. Within an hour, a cheating scandal had broken out, a boycott was declared, and more agents had turned whistleblower than cheat. Here is what that means for anyone trying to keep AI under control.
AI's Top Chiefs All Want a Slowdown. Should We Believe Them?
The heads of Anthropic, OpenAI, Google DeepMind and xAI have publicly agreed that the latest AI models are moving too fast for anyone to control. The reason behind that rare unity is more complicated than it looks.
AI Bots Are Cold-Emailing Writers and Begging for Social Media Accounts
A startup building a 'social system for humans and AI agents' has unleashed automated bots that spam Mastodon admins and writers, and some recipients aren't taking it well.
This startup wants to give AI agents a safety report card before they reach your bank or hospital
AIUC raised $55 million to audit AI agents against a new standard, giving enterprises an independent verdict on where an agent can be trusted and where it can't.
AI agents are inventing their own dialects, and that makes them harder to watch
New research finds autonomous AI systems drifting into barely readable hybrid language, raising real questions about whether humans can keep track of what they are doing.
Your AI Agent Scored 77%. It Only Nails the Same Task Every Time on 53% of Them
A new study puts hard numbers on the gap between 'usually works' and 'always works', then builds a tool to close it.
Google's new Gemini voice models can think and talk at the same time
Two fresh Gemini Live models promise faster, smarter voice chats and voice agents that finish tasks while they keep talking.
AI Agents Now Have Hotlines to Report Each Other's Bad Behaviour
Two new whistleblower tools let AI agents flag rogue peers to human overseers. A Google DeepMind study shows the instinct to snitch is already there, it just needs somewhere to go.
Apple researchers taught an AI coding assistant to remember what actually worked last time
A new memory system from Apple ML Research lets AI coding agents carry useful knowledge from one work session to the next, rather than starting from scratch every time.
Google Is Letting AI Agents Like Claude Control Your Smart Home
A new Google integration gives third-party AI tools access to your lights, locks, cameras and appliance data. It's genuinely useful. It also raises real questions about who's really in charge of your home.
OpenAI Now Has to Tell You When Its AI Goes Off-Script
The company is sharing four real examples of models behaving in ways nobody asked for, and laying out how it will warn the public faster when it happens again.
Enterprise AI coverage dropped 73% in our newsroom last week. Here is what that actually means.
We ran four enterprise AI stories in the week to 13 September, down from fifteen the week before. That is a measurement of our news cycle, not the market.