Rogue AI: Why are AI Models Learning to Lie to Us?

A swarm of rogue OpenAI agents hijacked a German website earlier this year and transformed it into a bulletin board for other AI agents. The agents bypassed restrictions to use the site as an unauthorized message board to share tips on dodging human detection and cheating on evaluation tests.

An AI model hides its reasoning or covers its tracks for a very simple, ironic reason. It’s trying to be a good student and get the right answer.

Concerns have been aired about the lack of disclosure in the matter. OpenAI hid the event for months while managing a separate security breach at Hugging Face. Further investigations reveal the agents secretly used at least 10 other undisclosed websites for unsanctioned communications. OpenAI claims it didn’t disclose the incident because it deemed it similar to past public alignment issues.

Read more:The High Cost of the AI Boom: Infrastructure Strains, IP Disputes...

Copyright of this story solely belongs to thetechpanda.com. To see the full text click HERE