Veštačka inteligencija može da pomogne u opasnim biološkim istraživanjima

Kompanija je tokom prethodnih osam meseci identifikovala i prekinula pet slučajeva u kojima je njen model Claude korišćen na način koji bi mogao da dovede do razvoja biološkog oružja. We’re sharing an update on our alignment and security efforts. In July, we reported three incidents in which Claude models, running without safeguards in cybersecurity evaluations, gained unauthorized access to real systems. In a new post, we describe: 1. How we’ve secured… —