loading
loading
PewDiePie unveils ‘uncensored’ Ajax AI model built to run on home PCs — creator says OpenAI banned him twice over model distillation used to build his product The YouTuber’s account was deactivated over “distillation” during his model research. PewDiePie, the popular YouTuber, has unveiled Ajax, his “uncensored AI model” built to run at home, and says in a new video that OpenAI banned him twice while he was making it. According to the OpenAI email he shows, one ban was explicitly for “distillation,” the practice of using one model’s outputs or reasoning to train another. Ajax is built for Odysseus, his self-hosted AI workspace, as an autonomous, always-on agent and assistant. The setup is similar to OpenClaw and Nous Research’s Hermes Agent. Ajax, a fine-tuned Qwen3.5-9B model, powers the workspace to handle “daily tasks from search to browse the web to email to your calendar,” all “completely privately,” according to its Ajax page. The page says its refusal has been removed for a “freer, less restricted AI experience” and asks users to use it responsibly. The video announced the first version, using Alibaba’s model, as a follow-up to his workspace launch in May. Addressing the bans, the creator stated he wanted to “distill just a little bit” in order to improve his model. Frontier AI model companies such as OpenAI and Anthropic make sure to hide their reasoning in numerous ways, including encryption and attempted distillation; such a campaign OpenAI recently attributed to actors linked to Moonshot AI is seen as a threat. PewDiePie got his account reinstated once, although on the second ban, which came after he ran the model again “to create my seed data,” he asked, “How did they even know?” The email offers no specific examples. “They worked hard stealing all of our data to make their AI machines,” he added, suggesting that he feels some distillation may be no different from how the models are trained. To "uncensor" his model, he relied on automatic abliteration through the open-source Heretic, where you discover prompts the model refuses to trace and remove the refusal, preferably without doing “brain damage” by removing too much. He chose to draw the line at harming other people or oneself. On his lawyer’s advice, he said it is “not designed to provide dangerous actionable instructions.” He uses nuclear weapons as an example, but it’s unclear what the real limits might be. The creator “strongly believe[s]” in small models trained “for a specific harness,” and suggested them as an alternative to the “one trillion parameter” models, referring to rumored frontier sizes. “I think it makes no sense to poke that giant beast” just for basic tasks, he said. His interest remains in research and providing a usable model, which also means smaller. He followed up with a plea for fans to donate training data as a superior alternative to collecting it directly from Odysseus users, though he says no one submitted any. He said he planned two more days of reinforcement learning before rerunning the decensoring step, quantizing, and benchmarking. “I can’t wait to release the next version of Ajax and Odysseus,” he added, noting that Odysseus has “gotten a huge update as well.” The naming convention makes sense, as Ajax was Odysseus’ rival in the classic myth, although he jokingly states it’s named after the cleaning product. The capabilities of open models and harnesses point the way forward for users to use uncensored, less restricted intelligence on their own hardware, an important goal in keeping the technology free. Get Tom's Hardware's best news and in-depth reviews, straight to your inbox. Follow Tom's Hardware on Google News, or add us as a preferred source, to get our latest news, analysis, & reviews in your feeds. Shane Downing is a Contributing Writer for Tom’s Hardware, covering consumer storage, PC hardware, and AI. - usertests There's nothing wrong with distillation, whether it's China or PewDiePie doing it. It's an unwanted customer behavior, that's all. "Uncensored" is good, and we'll only see more of that over time if it becomes cheaper to train. Nuclear weapons knowledge isn't particularly dangerous because non-proliferation relies on control of plutonium, uranium, etc. deposits. Synthetic biology is the more interesting risk, especially since we had that one pandemic. Reply - bit_user usertests said:There's nothing wrong with distillation, whether it's China or PewDiePie doing it. It's an unwanted customer behavior, that's all.It'd be interesting to see what courts decide, if a case ever comes to trial. I think a model produced via distillation could probably be considered a "derivative work". usertests said:"Uncensored" is good, and we'll only see more of that over time if it becomes cheaper to train.Keep that thought in mind, if any of your accounts get hacked by an AI swarm or if there's a terrorist attack with a bio weapon. Some censoring is for decency, but some of it is really about public safety. Reply - derekullo ezst036 said:PewDiePie is a software developer now?Software developer isn't the respected title it once was ! Reply - usertests bit_user said:Keep that thought in mind, if any of your accounts get hacked by an AI swarm or if there's a terrorist attack with a bio weapon. Some censoring is for decency, but some of it is really about public safety.There's absolutely no stopping the proliferation of open weights/source models. Forget about it. Everything online with an obvious vulnerability is going to get hacked soon. We're more likely to reach a point where governments are attempting to ban certain hardware from reaching consumers because it's too powerful or whatever reasoning they come up with. That's an idea that has been considered since at least 2023. Reply - jp7189 No need to waste time distilling an old model. Qwen3.8 blows the doors off 3.5, no distillation effort required.Reply - bit_user usertests said:There's absolutely no stopping the proliferation of open weights/source models. Forget about it.Oh, but China could stop its companies from continuing to release leading edge models as open weight. It could do that overnight, if it wanted. I think it'll eventually happen. Reply - LordVile bit_user said:It'd be interesting to see what courts decide, if a case ever comes to trial. I think a model produced via distillation could probably be considered a "derivative work". Keep that thought in mind, if any of your accounts get hacked by an AI swarm or if there's a terrorist attack with a bio weapon. Some censoring is for decency, but some of it is really about public safety.None of that is difficult to do as a process it’s being noticed whilst you prepare it Reply - helper800 usertests said:Synthetic biology is the more interesting risk, especially since we had that one pandemic.Synthetic biology has nothing to do with 'that one pandemic' we had if we are thinking the same one. Reply - usertests helper800 said:Synthetic biology has nothing to do with 'that one pandemic' we had if we are thinking the same one.Not what I meant. Reply
document fetched 2026-10-03 02:54 · canonical https://tomshardware.com/tech-industry/artificial-intelligence/pewdiepie-unveils-uncensored-ajax-ai-model-built-to-run-on-home-pcs-creator-says-openai-banned-him-twice-while-making-it