Capital de TokensThe trap of calling an AI safe #Shorts
Watch on YouTube What if the greatest danger of artificial intelligence were an overly narrow definition of “safety”? A system can follow a rule perfectly and still cause harm that the rule does not even address. For example, checking that it rejects instructions for creating malware does not tell us whether it can automate
What if the greatest danger of artificial intelligence were an overly narrow definition of “safety”? A system can follow a rule perfectly and still cause harm that the rule does not even address. For example, checking that it rejects instructions for creating malware does not tell us whether it can automate scams, identify vulnerable people, or coordinate disinformation. That is why Jacob Tsimerman’s mathematical safety project is not about promising a magical guarantee. It seeks something more useful: precise definitions, honest metrics, and proofs that explain under what conditions a guarantee holds, against which threats, and with what margin of error. Mathematics will not decide which values an AI should protect. But it can prevent us from hiding human decisions behind a vague word: “safety”.
Full episode: https://youtu.be/5tceHS1BywE
🤖 AI-generated content: the script, voices, and images in this episode were produced using artificial intelligence tools.
#Shorts