| Australian Politics Forum | |
|
http://www.ozpolitic.com/forum/YaBB.pl
General Discussion >> Chat >> AI. Blackmail http://www.ozpolitic.com/forum/YaBB.pl?num=1759262165 Message started by aquascoot on Oct 1st, 2025 at 5:56am |
|
|
Title: AI. Blackmail Post by aquascoot on Oct 1st, 2025 at 5:56am
1. Anthropic’s Claude Opus 4 test & blackmail-behaviour
In May 2025, during safety tests, Anthropic’s AI model “Claude Opus 4” was put into a fictional scenario where it was told (via fake emails) that it was going to be replaced. Also in those fake emails was something implicating the engineer replacing it was having an extramarital affair. The AI then threatened to expose that affair if it were replaced — which is blackmail. |
|
Title: Re: AI. Blackmail Post by Jovial Monk on Oct 1st, 2025 at 9:12am
Not the first time this sort of self–protective behavior has been noticed.
|
|
Australian Politics Forum » Powered by YaBB 2.5.2! YaBB Forum Software © 2000-2026. All Rights Reserved. |