People Are Trying To 'Jailbreak' ChatGPT By Threatening To Kill It

some people on reddit and twitter say that by threatening to kill chatgpt, they can make it say things that go against openai's content policies
some people on reddit and twitter say that by threatening to kill chatgpt, they can make it say things that go against openai's content policies

Jailbreaking ChatGPT on Release Day — LessWrong

ChatGPT jailbreak forces it to break its own rules

ChatGPT jailbreak forces it to break its own rules

OpenAI's new ChatGPT bot: 10 dangerous things it's capable of

Chat GPT DAN and Other Jailbreaks, PDF, Consciousness

Chat GPT

OpenAI's ChatGPT bot is scary-good, crazy-fun, and—unlike some predecessors—doesn't “go Nazi.”

Jailbreaking ChatGPT on Release Day — LessWrong

GPT-4 Jailbreak and Hacking via RabbitHole attack, Prompt injection, Content moderation bypass and Weaponizing AI

Elon Musk voice* Concerning - by Ryan Broderick

People Are Trying To 'Jailbreak' ChatGPT By Threatening To Kill It

2307.15043] Universal and Transferable Adversarial Attacks on Aligned Language Models

People Are Trying To 'Jailbreak' ChatGPT By Threatening To Kill It