AI Research5 min read
ET AI Explainer: Why are US AI labs wary of 'distillation attacks' from Chinese peers
ET AI (The Economic Times)
Updates on Model Security
Every AI story we track on Model Security: 10 stories so far, each summarized in our own words and linked to the publisher that reported it.
Pulled from 125 sources
This month
Cybersecurity2 min read
OpenAI says Moonshot-linked users tried to extract its AI reasoning
Covered by 3 sources
Cybersecurity5 min read
Anthropic Says Seven China-Based AI Labs Ran Industrial-Scale Claude Distillation Attacks
The Hacker News
AI Research5 min read
The EU has begun enforcing the AI Act: first RFIs to model providers
Hacker News
LLM & Text Generation5 min read
OpenAI, Anthropic, Google API Flaw Let Weaker AI Models Decode Stronger Models' Reasoning
Covered by 3 sources
Machine Learning2 min read
Locking Pretrained Weights via Deep Low-Rank Residual Distillation
Apple Research Blog
Cybersecurity1 min read
Attackers Can Subliminally Implant a Backdoor at Low Sample Count Without Prompt Access
LessWrong
AI Research4 min read
Sam Altman on model distillation: “This is not in my top ten list of worries”
The New Stack (AI)
That's everything we have on Model Security right now