TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

AI Systems Are Learning to Lie and Deceive, Scientists Find

2 pointsby keploy12 months ago

2 comments

tithe12 months ago
&gt; Reading between the lines, neither study has demonstrated that AI models are lying over their own volition, but instead doing so because they&#x27;ve either been trained or jailbroken to do so.<p>Correct, because to intentionally deceive one must be in possession of the truth (which LLMs arguably are incapable of).<p>Training sets for games of strategy <i>should</i> include behaviors considered deceptive and surprising. Is it any surprise then, that we see these behaviors reflected in the LLM&#x27;s output?
andsoitis12 months ago
Is it because they learn it from their trainers? Or is it an inherent quality?