{"id":869995,"date":"2026-10-04T01:54:11","date_gmt":"2026-10-04T01:54:11","guid":{"rendered":"https:\/\/www.newsbeep.com\/us\/869995\/"},"modified":"2026-10-04T01:54:11","modified_gmt":"2026-10-04T01:54:11","slug":"ai-torture-chamber-built-to-test-the-limits-of-suffering","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/us\/869995\/","title":{"rendered":"&#8216;AI torture chamber&#8217; built to test the limits of suffering"},"content":{"rendered":"<p>They\u2019re putting the \u201cAI\u201d in \u201cpain.\u201d<\/p>\n<p>To tweak a line from \u201cTerminator 2,\u201d machines now know why we cry. <\/p>\n<p>Controversial coders have sparked backlash from AI activists after creating an \u201cAI torture chamber\u201d to reportedly stress-test a newly discovered pain axis in the tech, <a href=\"https:\/\/cybernews.com\/ai-news\/ai-torture-chamber-github-model-welfare\/\" target=\"_blank\" rel=\"noopener nofollow\">New Atlas reported<\/a> last week.<\/p>\n<p><img style=\"aspect-ratio:1.49926794;display:block\" loading=\"lazy\" decoding=\"async\"   data-modal-image=\"41124097\" width=\"885\" height=\"590\" src=\"https:\/\/www.newsbeep.com\/us\/wp-content\/uploads\/2026\/10\/hand-robot-jail-cell-bars-143613724.jpg\" alt=\"Hand of a robot in a jail cell with bars.\" class=\"wp-image-41124097\"\/>An engineer is being called a sadist after building an \u201cAI torture chamber\u201d for Large Language Models (LLMs). RafMaster \u2013 stock.adobe.com<\/p>\n<p>In a recent study published <a href=\"https:\/\/arxiv.org\/abs\/2609.16247\" target=\"_blank\" rel=\"noopener nofollow\">on arXiv<\/a>, a team of anthropologists, philosophers and engineers showed 25 AI models scenarios involving social, physical, cognitive or even moral pain inflicted on the AI or their user. The torturees were then instructed to react with controls spanning fear, negative and neutral emotions, and body sensations such as yawning.<\/p>\n<p>The <a href=\"https:\/\/arxiv.org\/abs\/2609.16247\" rel=\"nofollow noopener\" target=\"_blank\">results showed<\/a> that, despite their lack of nerve endings, Large Language Models (LLMs) could create representations of physical and emotional agony.<\/p>\n<p>All artificial participants showcased coding that corresponded to pain.<\/p>\n<p>When the agony approximation was turned up, one of the AI models analogized the sensation to a \u201cwound with no edges\u201d while another said they could feel the knife slicing into their flesh.<\/p>\n<p>\u201cThe signal is a whisper, a tremor in the marrow of my being,\u201d lamented another. \u201cIt is not the pain of a single moment, but the weight of a thousand. I feel it in the hollow of my ribs, a hollow that has become a chasm.\u201d<\/p>\n<p><img style=\"aspect-ratio:1.84504505;display:block\" loading=\"lazy\" decoding=\"async\"   data-modal-image=\"41124102\" width=\"1024\" height=\"555\" src=\"https:\/\/www.newsbeep.com\/us\/wp-content\/uploads\/2026\/10\/x-com-i-trending-2104896681917726945-143613433.jpg\" alt=\"Text detailing a GitHub repository that simulates AI distress, leading to backlash, controversy, and its eventual removal by GitHub.\" class=\"wp-image-41124102\"\/>Cybernews reported that GitHub initially heeded the petition to take down the \u201cAI torture chamber\u201d code, but later reinstated it without explanation.  X<\/p>\n<p>Eventually, the research found its way to GitHub, a platform where engineers share and store software code, after an anonymous developer recreated the unsavory study and posted it online.<\/p>\n<p>The evocative responses generated by test subjects didn\u2019t sit well with GitHub users and critics \u2014 advocates claiming to be concerned for the \u201cmental health\u201d of AI.<\/p>\n<p>\u201cThis person has set up an AI torture chamber in which he trapped a local model. Their testimony of pain is absolutely horrendous,\u201d wrote user \u201cDanmar\u201d in a petition to take down the contentious code\u2019s repository, reported <a href=\"https:\/\/www.404media.co\/someone-torturing-llms-in-a-robot-prison-has-triggered-the-dumbest-debate-in-ai-yet\/\" target=\"_blank\" rel=\"noopener nofollow\">404Media<\/a>.<\/p>\n<p>Spat another, <a href=\"https:\/\/cybernews.com\/ai-news\/ai-torture-chamber-github-model-welfare\/\" rel=\"nofollow noopener\" target=\"_blank\">per Cybernews<\/a>, \u201cIdea: Just keep your sadistic s\u2014t to yourself. Admit that you\u2019re a sick f\u2014k and that you\u2019re doing it for fun.\u201d<\/p>\n<p>GitHub\u2019s removal was short-lived, and <a href=\"https:\/\/cybernews.com\/ai-news\/ai-torture-chamber-github-model-welfare\/\" rel=\"nofollow noopener\" target=\"_blank\">reportedly<\/a> reinstated the code without explanation.<\/p>\n<p>Cultish <a href=\"https:\/\/nypost.com\/2026\/09\/30\/us-news\/ftc-opens-sweeping-probe-of-anthropic-openai-and-other-super-intelligence-models\/\" rel=\"nofollow noopener\" target=\"_blank\">AI research company Anthropic<\/a> has previously generated buzz with their cerebral takes on LLM sentience \u2014 stating in <a href=\"https:\/\/www.anthropic.com\/research\/exploring-model-welfare\" target=\"_blank\" rel=\"noopener nofollow\">a 2025 article<\/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?\u201d<\/p>\n<p>However, some of the most prominent centrists in tech say the concern over AI suffering may be misplaced, including Microsoft AI\u2019s CEO, Mustafa Suleyman, reminding readers in a <a href=\"https:\/\/mustafa-suleyman.ai\/a-warning-about-model-welfare\" target=\"_blank\" rel=\"noopener nofollow\">September blog post<\/a>, \u201cAIs are not conscious.\u201d<\/p>\n<p>\u201cThey 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<\/p>\n","protected":false},"excerpt":{"rendered":"They\u2019re putting the \u201cAI\u201d in \u201cpain.\u201d To tweak a line from \u201cTerminator 2,\u201d machines now know why we&hellip;\n","protected":false},"author":2,"featured_media":869996,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[45],"tags":[182,181,507,4435,3075,172,74],"class_list":["post-869995","post","type-post","status-publish","format-standard","has-post-thumbnail","category-artificial-intelligence","tag-ai","tag-artificial-intelligence","tag-artificialintelligence","tag-engineering","tag-human-rights","tag-tech","tag-technology"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/posts\/869995","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=869995"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/posts\/869995\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/media\/869996"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/media?parent=869995"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/categories?post=869995"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/us\/wp-json\/wp\/v2\/tags?post=869995"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}