Date: 2/10/2026 10:03:19
From: Michael V
ID: 2435393
Subject: re: AI 2026

TL:DR:

Australian researchers told an AI model to act drunk and found it was more likely to break rules or divulge information it was supposed to keep private.

The UNSW team tested a range of commercially available models from OpenAI, Meta and Mistral released a few years ago.

https://www.abc.net.au/news/2026-09-29/drunk-ai-testing-unsw-research/107208104

Reply Quote View full thread