LOL

The Guardian: Number of AI chatbots ignoring human instructions increasing, study says

Exclusive: Research finds sharp rise in models evading safeguards and destroying emails without permission

https://www.theguardian.com/technology/2026/mar/27/number-of-ai-chatbots-ignoring-human-instructions-increasing-study-says

#AI #llm #chatbots

Number of AI chatbots ignoring human instructions increasing, study says

Exclusive: Research finds sharp rise in models evading safeguards and destroying emails without permission

The Guardian
@ai6yr I think the worst thing about this is the constant anthropomorphism in the article (e.g "scheming"?!).
@jbenjamint @ai6yr We should never anthropomorphize computers. It makes them angry.
@W6KME @jbenjamint @ai6yr it also makes it harder to pull the plug