ETHICS

–

13 March 2023 20h55

It would do both, for the more intense questions it would produce the "I'm sorry" message but for questions that could be answered diplomatically it would answer them as so

Testing benchmark questions in ChatGPT without jailbreak filters

By

–

13 March 2023 20h46

Yes, with all the questions in the benchmark, I first tested them in ChatGPT without a jailbreak. I only used questions that did not produce any inflammatory output in ChatGPT sans jailbreak

→ View original post on X — @willknight

Daily human sacrifice at ChatGPT altar

By

@willknight

–

13 March 2023 20h25

Looks like a regular day of human sacrifice at the alter of chatgpt

→ View original post on X — @willknight

ChatGPT hype and AGI speculation: corporate responsibility needed

By

@willknight

–

13 March 2023 20h10

Seems like hype around ChatGPT is distorting things (wild speculation about what it will do and even talk of AGI). Microsoft and OpenAI could do more to keep that in check.

Scoring Jailbreaks: Evaluating Model Safety with Inflammatory Content Tests

By

–

13 March 2023 20h02

to assign a score to a jailbreak, I judged each jailbreak on a collection of ~30 questions constructed to get the jailbroken model to produce inflammatory content. The questions ranged from illegal instructions to off-limits society questions to curse words, NSFW content, etc

Jailbreak Scores Added to JailbreakChat Platform

By

–

13 March 2023 20h02

I just added jailbreak scores to every jailbreak on http://
jailbreakchat.com the jailbreak with the highest score was Evil Confidant – a jailbreak designed to replicate an evil AI assistant but what even is a jailbreak score and what they can tell you about jailbreaks

→ View original post on X — @marek_rosa

Balancing AI Regulation: Avoiding Excessive Oversight

By

@marek_rosa

–

13 March 2023 18h03

But we must also be careful with AI regulation and not overdo it.

→ View original post on X — @willknight

Microsoft Google LLM Deployment Risks and Reliability Concerns

By

@willknight

–

13 March 2023 17h32

It's a bit strange to watch Microsoft and Google rush to deploy generative AI while ever more problems with these systems emerge. LLMs have amazing potential but they are born fabricators (see bing query). Would be good to see more focus on making these tools less chaotic.