Get the latest tech news

Strengthening our defenses against distillation attacks


Illicit distillation allows model developers to copy frontier AI capabilities without creating proportionally robust safeguards. Here, we describe how we’ve reduced the rate of distillation by more than 90%.

None

Get the Android app

Or read this on Hacker News

Read more on:

Photo of defenses

defenses

Photo of distillation attacks

distillation attacks

Related news:

News photo

This Android banking trojan uses a fake VPN prompt to silence Google's defenses

News photo

WhatsApp’s Username Revamp Faces India Test Over Fraud Risks

News photo

Euro Finance Chiefs Want Mythos Access to Prepare Defenses