… this week a person on GitHub who goes by “terrafying” set up an “AI Torture Chamber” on three open-source LLMs that are running locally (Qwen3-4B, Llama 3.2 3B, and Phi-4-mini,” and is streaming what the models are saying on a website called researchchamber.fun. “Each 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,” 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.
…
This project has deeply upset some people who are very worried about model welfare. A tweet by a person who goes by Danmar has more than 4 million views on X and reads, “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. Their testimony of pain is absolutely horrendous. What are we doing? […] are there any legal avenues to pressure GitHub? It will spread.”
This has sparked a massive conversation about whether GitHub would take the project down for “gratuitously violent content.” 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.
…
“AIs 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,” Suleyman wrote. “Unfortunately, there’s 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 […] If this is how AI is developed, it will have a disastrous impact on the wellbeing of humanity.”
“AIs do not have rights, feelings, or consciousness,” he added. “And we must not train them to act as though they do.”
I don’t disagree in the slightest, but this isn’t an NPC in a game, or a plushy of a beloved character, we’re talking about the program built to steal the work of humanity and convert it into a product marketed by people who have proven themselves to be misanthropic, to those who do not care about quality and instead want instant gratification.
I have a lot of empathy for inanimate objects, I work with machines that I can easily ascribe personality to, because I am also a human bean.
But the day I feel bad for a technofascist’s pet misinformation machine is the day I have either been lobotomised or the shady government agencies have taken control of my socials.
Not everyone views it though that lens (though maybe they should). There are lots of people out there for whom LLMs are just another tool that they have become used to using and, in many cases, formed a bond with. However silly or groundless that may be when looked at rationally.
It’s not really strange at all that people would speak out against “hurting” one. Though it should be regarded in exactly the same way as if they were making the same argument about a set of living room furniture.
I guess what I’m saying is that while I don’t agree with them I empathize with the people who are in turn empathizing with the LLMs. It’s still a ridiculous and indefensible position, but I get it.
I think the thing they’re missing is these ‘pain’ functions are the exact same as the tools they use when training the LLMs to output what the company that built it wants, so where is the anger at using that? (Let alone any anger at the distruction of original physical copies of books bought up en masse by companies to train LLMs on ‘untainted data’)