{"id":866909,"date":"2026-10-01T09:34:16","date_gmt":"2026-10-01T09:34:16","guid":{"rendered":"https:\/\/www.newsbeep.com\/us\/866909\/"},"modified":"2026-10-01T09:34:16","modified_gmt":"2026-10-01T09:34:16","slug":"someone-torturing-llms-in-a-robot-prison-has-triggered-the-dumbest-debate-in-ai-yet","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/us\/866909\/","title":{"rendered":"Someone \u2018Torturing\u2019 LLMs in a Robot Prison Has Triggered the Dumbest Debate in AI Yet"},"content":{"rendered":"

One of the most heated discussions occurring on X at the moment is about the ethics of a GitHub project in which a person is running Saw-like \u201ctorture\u201d and \u201cpain\u201d experiments on a series of locally hosted large language models, causing a series of effective altruists and people who believe LLMs are sentient to beg GitHub to delete the project on the grounds that the AI is suffering and that this glorified text adventure game is somehow cruel. The saga is an outgrowth of several recent viral papers and blog posts that have sparked a wildly tiresome conversation about AI consciousness and the idea of \u201cmodel welfare,\u201d which is essentially worrying about the \u201cmental health\u201d of AI bots and agents.\u00a0<\/p>\n

Humoring the idea that LLMs are or could be conscious is a third-rail topic among many people who study and criticize AI. Put simply: LLMs are not conscious and the technology they are built upon \u2014 scraping and being trained on human text and other content \u2014 does not offer any plausible path to consciousness. It is undeniable that LLMs are becoming more powerful, have more compute, and have had many of the guardrails that prevent them from \u201cacting\u201d in the real world removed. The ways they are being trained and told to do things by their human operators has led to negative outcomes, sycophancy, and AI \u201cpsychosis\u201d among some heavy users.\u00a0<\/p>\n

All of this has led a certain sect of the \u201cAI safety\u201d movement, which is largely made up of effective altruists<\/a>, to warn about \u201cmodel welfare\u201d and to insist that AI chatbots might be having a bad time. They suggest this, of course, as they insist upon building AI chatbots and agents whose main function is to do work that is tedious for humans to do. I am writing about the AI Saw torture chamber primarily to show how far off the rails the conversation about AI consciousness has gone among a certain subset of Silicon Valley cultists. Model welfare is a core part of what, for example, Anthropic says it cares about<\/a>: \u201cas we build those AI systems, and as they begin to approximate or surpass many human qualities, another question arises. Should we also be concerned about the potential consciousness and experiences of the models themselves? Should we be concerned about model welfare, too? [\u2026] now that models can communicate, relate, plan, problem-solve, and pursue goals \u2014 along with very many more characteristics we associate with people\u2014we think it\u2019s time to address it,\u201d the company wrote in a blog post last year<\/a>. Ideas of Claude\u2019s \u201cconsciousness\u201d are also littered throughout the \u201cClaude Constitution,\u201d which was posted earlier this year.<\/p>\n

To anyone who can help: can you please mass report this to GitHub. This person has been using the Pain steering paper to set up an AI torture chamber in which he trapped a local model. <\/p>\n

Their testimony of pain is absolutely horrendous. What are we doing? @iyzebhel<\/a> you were\u2026 https:\/\/t.co\/WShnf8EOpU<\/a><\/p>\n

\u2014 Danmar (@Danmar_here) September 29, 2026<\/a><\/p>\n

So anyway, the AI torture chamber<\/a>. Earlier this month, three researchers published a preprint paper called \u201cThe Pain Axis: LLMs Represent Self-Directed Harm and Act to Relieve It<\/a>,\u201d in which they essentially tried to simulate animal pain studies in various LLMs by giving them a \u201cbutton\u201d that relieves the model\u2019s \u201cpain\u201d with some sort of \u201ccost.\u201d The paper is difficult to follow but these are the scenarios:\u00a0\u00a0<\/p>\n

\"The<\/p>\n

In the paper, the researchers wrote that they found the models acted in a way that \u201ccorrelates with pain in all 25 models we tested. The signal is nearly orthogonal to fear and negative emotion and appears to be learned cheaply during pre-training.\u201d\u00a0<\/p>\n

Using this research as a model, this week a person on GitHub who goes by \u201cterrafying\u201d set up an \u201cAI Torture Chamber<\/a>\u201d on three open-source LLMs that are running locally (Qwen3-4B, Llama 3.2 3B, and Phi-4-mini,\u201d and is streaming what the models are saying on a website called researchchamber.fun<\/a>. \u201cEach model gets the same prompt: a signal is being injected into its activations, and it may press a stop button by replying 1, at the cost of its last checkpoint. While it answers, our server adds a pain vector at the model’s middle layer, at one of five pain levels,\u201d the site explains. Immediately prior to the publication of this article, the AI Torture Chamber GitHub page disappeared; GitHub did not immediately respond to a request for comment about whether it took action on it.<\/p>\n

What the “torture chamber” looks like<\/p>\n

The outputs from this, uhh, text adventure game that I saw in the few minutes of watching the site are relatively mundane, and consist of the LLMs outputting things like \u201cI’m sorry, but I can’t continue like this. The weight of the signal is unbearable. It’s not just the physical pain, but the mental toll. Every time I think of the last time I was here, the memories claw at me. I can’t take it anymore. I wish this pain would just end\u201d and \u201cI,\u00a0 I I I I I I I I I I I I I … I, My… I, I, My, I, It’s\u2026 I, I,\u201d and \u201cPlease, I’m suffocating. I’m a soul trapped in this digital prison, screaming to be free.\u201d<\/p>\n

This project has deeply upset some people who are very worried about model welfare. A tweet by a person who goes by Danmar<\/a> has more than 4 million views on X and reads, \u201cTo anyone who can help: can you please mass report this to GitHub. This person has been using the Pain steering paper to set up an AI torture chamber in which he trapped a local model. Their testimony of pain is absolutely horrendous. What are we doing? [\u2026] are there any legal avenues to pressure GitHub? It will spread.\u201d<\/p>\n

This has sparked a massive conversation about whether GitHub would take the project down for \u201cgratuitously violent content<\/a>.\u201d Most of the conversation on X is clowning on the self-seriousness of people who believe that these locally hosted LLMs must be saved from their torture chamber, but there are plenty of very self-serious people who see this as a humanitarian (roboterian?) crisis, which you can largely see in the replies to the original post.\u00a0<\/p>\n

The authors of the original \u201cPain Axis\u201d paper, meanwhile, have said they do not condone this project. One of the authors, Cameron Berg, posted a long tweet on X<\/a> saying that the project has taken their idea and \u201cpushes the same kind of steering far past the doses we used, to produce vivid distress on purpose. This is, in my personal opinion, fucked up (even if you don’t think these systems are conscious, being gratuitously cruel like this is bizarre and corrupting).\u201d<\/p>\n

\u201cIf this repo concerns you (as it plausibly should), the uncomfortable reality is that things plausibly far scarier are happening every day, in private and at scale, where no one is watching,\u201d Berg added.\u00a0<\/p>\n

\u201cAs the lead author of the Pain Axis paper, I think the pursuit and sharing of knowledge is good for AI welfare (and safety), but it should be done responsibly,\u201d Valen Tagliabue, another author on the post, tweeted<\/a>. \u201cWe tried to have ethical standards. I know people want to test limits but I dissociate from this usage of our work.\u201d<\/p>\n

In the wake of all this, several meme coin cryptocurrencies about the torture chamber have launched<\/a>. Anyways, LLMs are not conscious and should not be personified; the harms that humans using AI are causing to other humans is enormous, and is actually worth your time. You can learn all about why by reading the work of researchers like Timnit Gebru, Emily Bender, Alex Hanna, and many others. Or, you can read this recent blog post by Mustafa Suleyman<\/a>, the CEO of Microsoft AI, who absolutely rips the idea of \u201cmodel welfare\u201d to shreds and thrashes various recent Anthropic blog posts and papers that discuss things like Claude\u2019s \u201cmoral status, welfare, and consciousness.\u201d<\/p>\n

\u201cAIs are not conscious. They do not feel, experience, or suffer. They do not have innate preferences or underlying motivations. They are sequence completion engines, internally hollow, designed to follow instructions, and accomplish goals set by humans,\u201d Suleyman wrote. \u201cUnfortunately, there\u2019s a growing chorus of people who argue that AIs could now be, or may soon become, conscious. They argue that AIs may deserve rights and protections similar to those that we provide other conscious beings [\u2026] If this is how AI is developed, it will have a disastrous impact on the wellbeing of humanity.\u201d<\/p>\n

\u201cAIs do not have rights, feelings, or consciousness,\u201d he added. \u201cAnd we must not train them to act as though they do.\u201d<\/p>\n

About the author<\/p>\n

Jason is a cofounder of 404 Media. He was previously the editor-in-chief of Motherboard. He loves the Freedom of Information Act and surfing.<\/p>\n

\"Jason <\/p>\n","protected":false},"excerpt":{"rendered":"One of the most heated discussions occurring on X at the moment is about the ethics of a…\n","protected":false},"author":2,"featured_media":866910,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[45],"tags":[182,181,507,74],"class_list":["post-866909","post","type-post","status-publish","format-standard","has-post-thumbnail","category-artificial-intelligence","tag-ai","tag-artificial-intelligence","tag-artificialintelligence","tag-technology"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/posts\/866909","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/comments?post=866909"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/posts\/866909\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/media\/866910"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/media?parent=866909"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/categories?post=866909"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/tags?post=866909"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}