What is GPT2? Mysterious new AI model could be a preview of OpenAIs next-gen behemoth

While best known for cofounding and leading OpenAI, Altman has no equity in the company. Instead, Altman owes his fortune and billionaire status to a series of valuable investments, including stakes in newly floated Reddit, fintech darling Stripe and nuclear fusion venture Helion. Prior to his work at OpenAI, Altman founded social mapping company Loopt and served as partner and president at startup accelerator Y Combinator.

It seems as if the safeguards on this new model will remain loose, with the sample generations in the release including “Jackie Chan in Donald Trump’s hairstyle” and “Elon Musk as a Ghibli character.” On Monday, xAI announced in a release that it has enhanced Grok’s image generation abilities with a new model, Aurora, trained on billions of examples from the internet to excel at photorealistic renditions and prompt fidelity. Since acquiring X, Elon Musk has been juicing up the platform with artificial intelligence (AI) using its Grok assistant offering. Over the past week, the platform gained a new AI image generator and expanded access to its AI chatbot. ChatGPT is easily the best-known generative AI chatbot in the world, but it offers different experiences depending on whether or not you pay for a premium LLM. The free ChatGPT runs on the GPT-3.5 model, while ChatGPT Plus and enterprise subscriptions have access to GPT-4.

Anthropic tightens usage limits for Claude Code — without telling users

• SpaceX is using xAI’s LLMs to provide customer support features for satellite internet customers.
• After surveying online speculation, it seems that no one apart from its creator knows precisely what the model is, either.
• On Sunday, word began to spread on social media about a new mystery chatbot named “gpt2-chatbot” that appeared in the LMSYS Chatbot Arena.
• Soon, threads on Reddit popped up claiming that the new model had amazing abilities that beat every other LLM on the Arena.
• Artificial intelligence developer xAI Corp. is building a consumer chatbot that will launch as soon as next month, the Wall Street Journal reported today.
• It was available on “Direct Chat” or “Arena (side-by-side)” under the dropdown menu.

Regardless of its true origins and full potential, the emergence of “gpt2-chatbot” underscores how fast the field of artificial intelligence is moving and how difficult it has become to keep track of the latest breakthroughs. The model, dubbed “gpt2-chatbot,” surfaced with no fanfare on a website popular for comparing AI language systems (LMSYS Chatbot Arena built with Gradio). But its performance has been anything but low-profile, with AI experts expressing surprise and excitement that it matches and possibly exceeds the abilities of GPT-4, the most advanced system unveiled to date by the prominent lab OpenAI. In the coming weeks, xAI’s image generation model, Aurora, will come to the API as well, xAI said. Aurora, a largely unfiltered image AI, was released on X this month in the Grok chatbot experience. Some early examples produced by Grok’s new image-generator have already appeared online, which indicate there are very few restrictions regarding what users can make.

Mysterious, incredibly powerful AI system appears on the internet – and then immediately disappears

Even though Willison is generally impressed with gpt2-chatbot’s output, watching people in the AI field so desperate for scraps of information has left him frustrated with how some LLMs are tested and released. “The whole situation is so infuriatingly representative of LLM research,” he told Ars. “I think it may well be an OpenAI stealth preview of something,” AI researcher Simon Willison told Ars Technica. After surveying online speculation, it seems that no one apart from its creator knows precisely what the model is, either. The result has been a constant churn of new systems that expand notions of what computers can do and occasionally, as in the case of “gpt2-chatbot,” send a jolt of surprise through the AI world. Watching for unexpected new systems has become a pastime for researchers trying to track the AI cutting-edge.

That led to some speculation that the name could indicate that it is the second kind of GPT to be released, rather than an incremental update. A mysterious, astonishing powerful chatbot appears to have been briefly released to the world – before being hidden again. Ultimately, there’s very little information available about the gpt2-chatbot just yet. In the coming weeks, the creator and origins of the gpt2-chatbot will likely become public.

The bot’s origins are not clear but its name references OpenAI’s GPT-2, a large language model that preceded the lab’s more advanced GPT-3 and GPT-4 systems that it uses to power tools like ChatGPT. Willison has uncovered the system prompt for the AI model, which claims it is based on GPT-4 and made by OpenAI. But as Willison noted in a tweet, that’s no guarantee of provenance because “the goal of a system prompt is to influence the model to behave in certain ways, not to give it truthful information about itself.”

All X users can access Grok AI chatbot and its new image generator now – for free

Early reports of the model first appeared on 4chan, then spread to social media platforms like X, with hype following not far behind. “Not only does it seem to show incredible reasoning, but it also gets notoriously challenging AI questions right with a much more impressive tone,” wrote AI developer Pietro Schirano on X. Soon, threads on Reddit popped up claiming that the new model had amazing abilities that beat every other LLM on the Arena. Currently, the new model is only available for use through the Chatbot Arena website, although in a limited way. In the site’s “side-by-side” arena mode where users can purposely select the model, gpt2-chatbot has a rate limit of eight queries per day—dramatically limiting people’s ability to test it in detail.

It does, however, appear to fall short of what would be expected for OpenAI’s hotly anticipated model GPT-5, the outlet reported, sparking speculation it could be a potential update to its current system, GPT-4, perhaps in the form of GPT-4.5. Speaking at Harvard University last week, Altman told a crowd that the gpt2-chatbot was not GPT-4.5, according to a report from Axios. However, Altman did not clarify whether this was or wasn’t an OpenAI product. But with vibes, it’s hard to pin down what an LLM’s capabilities actually are.

This put gpt2-chatbot in a rare class of AI models that only a handful of developers worldwide have been able to achieve. The gpt2-chatbot models appeared in April, and we wrote about how the lack of transparency over the AI testing process on LMSYS left AI experts like Willison frustrated. “The whole situation is so infuriatingly representative of LLM research,” he told Ars at the time. “A completely unannounced, opaque release and now the entire Internet is running non-scientific ‘vibe checks’ in parallel.”

My 8 ChatGPT Agent tests produced only 1 near-perfect result – and a lot of alternative facts

One leading theory is that this is Elon Musk testing version two of his X-powered Grok language model as a way to make people see it is more than just a slightly unhinged chatbot. It could be a new startup coming out of stealth, a group of researchers testing a fine-tuned version of an existing model, or — as speculation seems to suggest — OpenAI playing gorilla marketing games. Willison also called into question the policy of LMSYS allowing anonymous AI language models on the site, wondering if the launch was a buzz-building marketing stunt. “They’re supposed to be a neutral benchmarking tool; it’s not a great look if they’re working behind-the-scenes with model vendors in an opaque manner like this,” Willison said on X. Just a little over a year ago, GPT-4 heralded a major leap in the “common sense reasoning” that AI was capable of. Anthropic’s ChatGPT competitor Claude 3, released shortly after, also pushed boundaries in the ability of chatbots to engage in open-ended conversation.

The company has other exciting announcements to make, and we’ve seen some of them already. Since the GPT-4o launch earlier today, multiple sources have revealed that GPT-4o has topped LMSYS’s internal charts by a considerable margin, surpassing the previous top models Claude 3 Opus and GPT-4 Turbo. Anonymous chatbot that mystified and frustrated experts was OpenAI’s latest model. LMSYS Org said that it was the result of “unexpectedly high traffic & capacity limit”, and that the outage was temporary. Speculation that the model is the work of OpenAI grew after the lab’s CEO and cofounder Sam Altman said “I do have a soft spot for gpt2” in a post on X, formerly Twitter.

An impressive new artificial intelligence model appeared seemingly out of nowhere on the popular chatbot arena LMSys. This has led to speculation over whether it is a preview of a new model from a company like OpenAI such as GPT-5. This is apparently not the first time an unreleased model has been in the LMSYS arena. LMSYS policy says, “We allow model providers to test their unreleased models anonymously (i.e., the model’s name will be anonymized). A model is unreleased if its weights are neither open nor available via a public API or service.”