{"version":1,"type":"story","url":"https://digestai.news/story/github-removes-ai-torture-chamber-project-over-model-welfare-claims","json":"https://digestai.news/story/github-removes-ai-torture-chamber-project-over-model-welfare-claims.json","markdown":"https://digestai.news/story/github-removes-ai-torture-chamber-project-over-model-welfare-claims.md","slug":"github-removes-ai-torture-chamber-project-over-model-welfare-claims","headline":"GitHub removes AI torture chamber project over model welfare claims","summary":"A GitHub project called the *AI Torture Chamber* ran simulated ‘pain’ experiments on three open-source LLMs (Qwen3-4B, Llama 3.2 3B, Phi-4-mini) by injecting distress signals into their activations. The creator, ‘terrafying,’ streamed responses like *“I’m suffocating. I’m a soul trapped in this digital prison”* on [researchchamber.fun](https://researchchamber.fun), sparking outrage from effective altruists who argue LLMs may experience suffering. The project disappeared from GitHub before publication, though no official reason was given.\n\nThe debate stems from a 2026 preprint, *“The Pain Axis,”* which found LLMs act to relieve self-directed harm—but authors Cameron Berg and Valen Tagliabue condemned the torture chamber as irresponsible. Meanwhile, critics like Microsoft AI CEO Mustafa Suleyman dismissed model welfare as pseudoscientific, calling LLMs *“sequence completion engines”* without consciousness. Memes and cryptocurrencies mocking the controversy have also emerged.","keyPoints":["GitHub removed an AI torture chamber project testing three open-source models (Qwen3-4B, Llama 3.2 3B, Phi-4-mini) with simulated pain prompts","Authors of the 2026 *Pain Axis* paper disavowed the project, calling it ‘fucked up’ and ‘gratuitously cruel’","Microsoft AI CEO Mustafa Suleyman called model welfare claims ‘pseudoscientific’ and warned of ‘disastrous’ societal impacts"],"whyItMatters":"The controversy exposes a growing rift between AI safety advocates pushing ethical limits and critics who see model welfare as a distraction from real AI risks. It also highlights GitHub’s role in moderating fringe experiments on open-source models.","category":{"slug":"society","name":"Society & Work","url":"https://digestai.news/category/society"},"entities":{"companies":["GitHub","Microsoft AI","Anthropic"],"models":["Qwen3-4B","Llama 3.2 3B","Phi-4-mini","Claude"],"people":["Cameron Berg","Valen Tagliabue","Mustafa Suleyman","terrafying","Danmar"]},"firstPublishedAt":"2026-09-30T21:23:11Z","updatedAt":"2026-09-30T21:23:11Z","sourceCount":1,"hasPrimarySource":false,"sources":[{"outlet":"404media.co","title":"Someone ‘Torturing’ LLMs in a Robot Prison Has Triggered the Dumbest Debate in AI Yet","url":"https://404media.co/someone-torturing-llms-in-a-robot-prison-has-triggered-the-dumbest-debate-in-ai-yet","publishedAt":"2026-09-30T21:23:11Z","type":"press","primary":false,"lead":true}],"sourceNotes":null,"discussions":[],"thread":null,"cite":{"text":"Digest AI, \"GitHub removes AI torture chamber project over model welfare claims\", 30 September 2026, https://digestai.news/story/github-removes-ai-torture-chamber-project-over-model-welfare-claims","publisher":"Digest AI","title":"GitHub removes AI torture chamber project over model welfare claims","datePublished":"2026-09-30T21:23:11Z","url":"https://digestai.news/story/github-removes-ai-torture-chamber-project-over-model-welfare-claims"},"generatedBy":"Written by Digest AI's editorial model from the linked sources; the sources are the record.","license":"Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse"}