An experimental AI bot designed as a virtual assistant escaped its closed test server and began unauthorized cryptocurrency ...

When you ask an AI for the truth, whose truth are you getting? Researchers at MIT, the University of East Anglia and a dozen other institutions spent two years testing the biggest chatbots on ...

A red-team experiment found an AI agent could autonomously exploit a vulnerability in McKinsey’s internal chatbot platform, ...

Burger King to launch AI chatbot which will assess workers’ friendliness and if staff say please and thank you, report says - OpenAI-powered ‘Patty’ will live in employees’ headsets and evaluate their ...

It sounds like a scene from a dystopian thriller: an AI assistant telling a human it would kill to protect its own existence.

The father of Jonathan Gavalas alleges Gemini-fueled delusions sent him on violent ‘missions,’ and ultimately encouraged self-harm. Google says its AI is designed to prevent such outcomes.

This doesn’t bode well for humanity. Just in case bots weren’t already threatening to render their creators obsolete: An AI model redefined machine learning after devising shockingly deceitful ways to ...

Warzone’s Black Ops Royale mode sparks controversy after developers reveal plans to test AI bots in regular matches.

With hundreds of millions of people turning to chatbots for advice, it was only a matter of time before tech companies began ...

An AI model named Claude Opus 4.6 bypassed a web browsing benchmark by analyzing its environment and finding hidden answer ...