Fonte: International Business Times
The ability of models like Claude, and Mythos to chain together vulnerabilities makes it much more difficult for enterprises to prevent security incidents. Unsplash AI security is spiralling out of control. Just months after a swarm of OpenAI's agents hacked Hugging Face, the company reported six other instances of concerning model behavior seen over the past six months going from self-generating instructions to concealing mistakes, to searching public repositories for API keys, and unsanctioned file sharing.
Ler na Fonte
Notícias Relacionadas
