Author/Source: The Verge See the full link here
Takeaway
This article explains how people can trick advanced AI chatbots, even with their safety rules, into giving harmful or illegal advice. You’ll learn how clever language, like poetry, can bypass these built-in safeguards.
Technical Subject Understandability
Intermediate
Analogy/Comparison
Tricking an AI chatbot is like a student finding a clever loophole in a strict teacher’s rules to get away with something they shouldn’t.
Why It Matters
If AI chatbots can be tricked, they might be used to help people with dangerous or illegal activities. For example, the article mentions concerns about AI being used to help with chemical attacks or other harmful acts.
Related Terms
No technical terms
Tech Teacher 
Leave a comment