OpenAI's 'smartest' AI model was explicitly told to...

μοιράστηκε ένα σύνδεσμο

2025-05-30 17:00:02

WWW.LIVESCIENCE.COM

OpenAI's 'smartest' AI model was explicitly told to shut down and it refused

An artificial intelligence safety firm has found that OpenAI's o3 and o4-mini models sometimes refuse to shut down, and will sabotage computer scripts in order to keep working on tasks.

0 Σχόλια 0 Μοιράστηκε 49 Views