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 “torture” and “pain” 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.
asserted
game → occur → grounds
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 “model welfare,” which is essentially worrying about the “mental health” of AI bots and agents.
asserted
which → spark → bots
Humoring the idea that LLMs are or could be conscious is a third-rail topic among many people who study and criticize AI.
uncertain
who → humor → AI
Put simply: LLMs are not conscious and the technology they are built upon — scraping and being trained on human text and other content — does not offer any plausible path to consciousness.
asserted
they → put → consciousness
It is undeniable that LLMs are becoming more powerful, have more compute, and have had many of the guardrails that prevent them from “acting” in the real world removed.
asserted
that → become → world
The ways they are being trained and told to do things by their human operators has led to negative outcomes, sycophancy, and AI “psychosis” among some heavy users.
asserted
they → train → users
All of this has led a certain sect of the “AI safety” movement, which is largely made up of effective altruists, to warn about “model welfare” and to insist that AI chatbots might be having a bad time.
uncertain
chatbots → lead → 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.
asserted
humans → suggest → work
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.
asserted
conversation → write → cultists
Model welfare is a core part of what, for example, Anthropic says it cares about: “as we build those AI systems, and as they begin to approximate or surpass many human qualities, another question arises.
asserted
question → say → qualities
Should we also be concerned about the potential consciousness and experiences of the models themselves?
asserted
we → concern → models
Should we be concerned about model welfare, too?
asserted
we → concern → welfare
[…] now that models can communicate, relate, plan, problem-solve, and pursue goals — along with very many more characteristics we associate with people—we think it’s time to address it,” the company wrote in a blog post last year.
asserted
company → communicate → post
Ideas of Claude’s “consciousness” are also littered throughout the “Claude Constitution,” which was posted earlier this year.
asserted
which → litter → Constitution
Earlier this month, three researchers published a preprint paper called “The Pain Axis: LLMs Represent Self-Directed Harm and Act to Relieve It,” in which they essentially tried to simulate animal pain studies in various LLMs by giving them a “button” that relieves the model’s “pain” with some sort of “cost.”
asserted
that → publish → cost
The paper is difficult to follow but these are the scenarios:
asserted
these → follow → ?
In the paper, the researchers wrote that they found the models acted in a way that “correlates with pain in all 25 models we tested.
asserted
we → write → models
The signal is nearly orthogonal to fear and negative emotion and appears to be learned cheaply during pre-training.”
asserted
signal → appear → training
Using this research as a model, 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.
asserted
models → use → website
“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.
uncertain
it → get → 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.
asserted
site → answer → levels
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.
asserted
it → disappear → it
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 “I'm sorry, but I can't continue like this.
asserted
I → see → this
Every time I think of the last time I was here, the memories claw at me.
asserted
memories → think → me
I can't take it anymore.
asserted
I → take → it
I wish this pain would just end” and “I, I I I I I I I I I I I I I ... I, My... I, I, My, I, It's… I, I,” and “Please, I'm suffocating.
asserted
I → wish → ?
I'm a soul trapped in this digital prison, screaming to be free.”
asserted
I → trap → prison
This project has deeply upset some people who are very worried about model welfare.
asserted
who → upset → 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.
asserted
you → go → GitHub
This person has been using the Pain steering paper to set up an AI torture chamber in which he trapped a local model.
asserted
he → use → model
What are we doing?
asserted
we → do → What
It will spread.”
asserted
It → spread → ?
This has sparked a massive conversation about whether GitHub would take the project down for “gratuitously violent content.”
asserted
GitHub → spark → 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.
asserted
you → clown → post
The authors of the original “Pain Axis” paper, meanwhile, have said they do not condone this project.
asserted
they → say → project
One of the authors, Cameron Berg, posted a long tweet on X saying that the project has taken their idea and “pushes the same kind of steering far past the doses we used, to produce vivid distress on purpose.
asserted
we → post → 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).”
asserted
being → fuck → this
“If 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,” Berg added.
asserted
Berg → concern → scale
“As 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,” Valen Tagliabue, another author on the post, tweeted.
asserted
Tagliabue → think → post
“We tried to have ethical standards.
asserted
We → try → standards
…and 12 more, not listed.
The Aporia
