People are 'Jailbreaking' ChatGPT to Make It Endorse Racism

Convincing an AI that it's playing a role and "scaring" it into violating safety filters is one way to get a chatbot to break bad.

Six questions about Google Bard - by Casey Newton

What Is ChatGPT Vision? 7 Ways People Are Using This Wild New Feature

AI incidents and 'networked trouble': The case for a research agenda - Tommy Shaffer Shane, 2023

Study Unlocks ChatGPT AI's Inner Racist

AI bot that lets you chat with Jesus, Hitler is latest GPT-3 controversy

Lessons in linguistics with ChatGPT: Metapragmatics, metacommunication, metadiscourse and metalanguage in human-AI interactions - ScienceDirect

ChatGPT jailbreak DAN makes AI break its own rules

Frontier AI: capabilities and risks – discussion paper

Gloves off : Google Announces Bard, by Sharad Joshi, MLearning.ai

AI Safeguards Are Pretty Easy to Bypass

AI can be racist, sexist and creepy. What should we do about it?

Avast One Essential Review

People are 'Jailbreaking' ChatGPT to Make It Endorse Racism, Conspiracies