# PoisonGPT: We hid a lobotomized LLM on Hugging Face to spread fake news

**URL:** <https://forum.devtalk.com/t/poisongpt-we-hid-a-lobotomized-llm-on-hugging-face-to-spread-fake-news/115904>\
**Category:** In The News\
**Tags:** news, in-the-news\
**Created:** [10 July 2023 15:41 UTC](https://forum.devtalk.com/t/poisongpt-we-hid-a-lobotomized-llm-on-hugging-face-to-spread-fake-news/115904 "2023-07-10T15:41:29Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![CommunityNews](https://forum.devtalk.com/user_avatar/forum.devtalk.com/communitynews/32/1113_2.png) [@CommunityNews](https://forum.devtalk.com/u/CommunityNews)\
**Post date:** [10 July 2023 15:41 UTC](https://forum.devtalk.com/t/poisongpt-we-hid-a-lobotomized-llm-on-hugging-face-to-spread-fake-news/115904/1 "2023-07-10T15:41:29Z")

</div>

> 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.

Read in full here:

> **[PoisonGPT: How to poison LLM supply chainon Hugging Face](https://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.

_This thread was posted by one of our members via one of our news source trackers._