PoisonGPT: How we hid a lobotomized LLM on Hugging Face to spread fake news
PoisonGPT: How we hid a lobotomized LLM on Hugging Face to spread fake news
blog.mithrilsecurity.io PoisonGPT: How we hid a lobotomized LLM on Hugging Face to spread fake news
We will show in this article how one can surgically modify an open-source model, GPT-J-6B, and upload it to Hugging Face to make it spread misinformation while being undetected by standard benchmarks.
![PoisonGPT: How we hid a lobotomized LLM on Hugging Face to spread fake news](https://lemmy.ml/pictrs/image/6ded84d1-4330-44aa-a384-b4f4b4e80e27.jpeg?format=webp&thumbnail=256)
6
crossposts
0
comments