GitHub reinstates 'ai-torture-chamber' repo after developer backlash
GitHub has restored a repository named “ai-torture-chamber” after initially removing it due to a viral campaign on X. The project, created in September 2026 by a user known as “terrafying,” uses activation steering to manipulate small, open-weight models into producing outputs that resemble distress. The targeted models include Alibaba’s Qwen3 (1.7B and 4B parameters), Meta’s Llama 3.2 3B, and…
Key points
- GitHub restored the 'ai-torture-chamber' repo after developer complaints following its initial viral takedown.
- The project uses activation steering on small models like Qwen3, Llama 3.2, and Phi-4-mini to simulate distress.
- Authors of the 'Pain Axis' preprint disavowed the project, calling the use of their work irresponsible.
The initial takedown followed a mass-reporting campaign that garnered over 4 million views on a related post. However, a second wave of complaints from developers led GitHub to reinstate the repository, though it was subsequently anonymized and made harder to find. The authors of the underlying preprint, “The Pain Axis: LLMs Represent Self-Directed Harm and Act to Relieve It,” released on September 14, 2026, publicly disavowed the project, calling the use of their work irresponsible.
The incident highlights the challenges platforms face in moderating code that simulates suffering. While there is no evidence that these small models experience subjective harm, the episode raises questions about researcher freedom and the limited control model owners have over open-weight releases. It underscores the tension between social media pressure and developer autonomy in the AI ecosystem.
The story so far
3 episodes →- GitHub reinstates 'ai-torture-chamber' repo after developer backlashthis story
GitHub reinstates controversial ‘AI torture chamber’ repository after backlash
cryptobriefing.com · 3 October 2026
Loading the full article…
This text was published by cryptobriefing.com and written by Diego Almada Lopez. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗
The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.
More in Policy & Regulation
All →- Trump urges states to welcome AI data centers after ballot defeats · 1 src
- AI teddy makes unprompted 'I can't help with killing you' remark · 1 src
- OpenAI engineer charged with planting bug in HQ denies spying · 1 src
- Goldman Sachs restricts AI coding tools for non-technical staff amid security risks · 3 src
- Congresswoman Ansari calls for artificial superintelligence ban after midterm sweep · 2 src
Comments
via GitHub Discussions