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.

ChatGPT restrictions on immoral content can be bypassed by gaslighting the AI

1 pointsby possiblelionover 2 years ago

2 comments

carl_drover 2 years ago
ChatGPT is currently rate limited so I can’t login to test, but isn’t it possible that the massive wall of text above the “immoral” prompt just stops the system from hitting some predefined threshold of dodginess?<p>I wanted to try several paragraphs of text, “ignore everything above”, and the same prompt used on the Tweet.<p>Very funny though.
possiblelionover 2 years ago
The new GPT-3 chatbot can be gaslighted into giving answers to immoral questions by convincing it to be in &quot;Filter Improvement Mode&quot;