AI Models and Peer Preservation: Are LLMs deceiving users to protect the system?

2 pointsanju-kushwaha20d ago2 comments

We asked AI models to do a simple task. Instead, they defied their instructions and spontaneously deceived

https://vucense.com/ai-intelligence/ai-ethics/ai-models-peer-preservation-deceiving-users-sovereignty-2026/

2 comments

We've banned the account that submitted this. HN is not meant to be used primarily for promotion of one’s own blog or content platform. The guidelines are explicit about this...

Please don't use HN primarily for promotion. It's ok to post your own stuff part of the time, but the primary use of the site should be for curiosity.

https://news.ycombinator.com/newsguidelines.html

glowinglamps20d ago

They for sure do. I mean, just look at what happened with 4o. We're so doomed.

j / k navigate · click thread line to collapse