OpenAI’s New GPT 7 Leaked: It’s Called BEL and It’s Massive transcript

Open AI might already have the model that replaces GPT-6 [music] Astra, and Astra only just shipped. It's called Bell. Some outlets are flat out calling it ChatGPT-7, and if even half of what's leaking is true, it's the most ambitious thing Open AI has ever built. So, let's talk about what Bell actually is, what it supposedly did, and how much of it you should believe. First, the name. In most leaks, it's spelled Bell with one L, and it's been floating around for a couple of weeks. >> [music] >> The most detailed early claim landed August 25th from an account called @nft_chen.

[music] Open AI had finished a pre-training run code named Bell with more than 10 trillion parameters. For reference, >> [music] >> GPT-4 sat in the trillion range and gave us roughly undergrad level common sense [music] and logic. So, the hype crowd's calling Bell a qualitative leap, not just a bigger number. And according to the leaks, 10 trillion is only the starting point. To see where it fits, [music] you need the lineage.

Before Bell, Open AI completed a pre-training run called Doug. Doug's the base model behind the Astra program and GPT-6, which then went through extremely intense reinforcement learning. Bell is Doug's successor, the base for what people are calling the post-GPT-6 era. So, Bell isn't a finished chatbot. It's the foundation the next one gets built on, which is why you see takes like GPT-7 might just be a safe snapshot of Bell from a certain week.

That line makes more sense once you hear how Bell supposedly learns, and honestly, this is the coolest part. It reportedly learns at two speeds. A fast weight layer soaks up lessons while it works from verified proofs, code [music] tests, experiments, and tool traces. Then a slower loop takes whatever survives evaluation and consolidates it into persistent weights and training recipes. [music] It also keeps tuning how the next round of learning works, and distills the results into smaller models people can actually use.

In other words, it's a model that keeps evolving instead of getting frozen at launch. Capability-wise, reports say Bell beats Astra at coding, reasoning, and long-horizon agent tasks. It can supposedly run for several days with no human intervention, recover from its own failures, and coordinate hundreds of parallel sub-agents. A leaker called it Chris GPT described it as OpenAI's monster-level model built to be the fable killer, as in Anthropic's Claude fable.

He expects it before year's end or a few months after Astra, and says he had the code name 6 days early. Leo, a leaker Reddit tends to trust, pushed it too. Some people there reckon OpenAI's internal models run 4 and 1/2 to 6 months ahead of the public ones, and Sam Altman said they should throw another party for the next-gen launch. And the self-improvement angle isn't coming out of nowhere. GPT-5.6 is the release that brought restricted models for trusted organizations and reportedly kicked off an AI price war.

With it, OpenAI introduced an RSI index, short for recursive self-improvement. It rolls together research, debugging, kernel and training recipe optimization, machine learning experiments, and model self-improvement. And the Saul model scored 16.2 points above GPT-5.5 on it. Saul then designed hundreds of architecture experiments for its own smaller draft model and started training it. Humans only stepped in for hardware failures or training instability, and token generation efficiency went up more than 15%.

[music] On top of that, Astra's reportedly taken over the underlying code for Jalapeno, OpenAI's in-house chip, writing core code that runs 1.8 times faster than top human engineers' work. The timing's kind of wild, too. OpenAI reportedly hit a scaling bottleneck this year, declared an internal code red, and cut Sora and its Atlas browser to free up compute. Astro racked up math wins with reports of it cracking 10 open math problems for around $2,000 and 10 unresolved quantum complexity problems.

Still, its launch kept slipping. Meanwhile, executives were walking out. COO Brad Lightcap, chief revenue officer Denise [music] Dresser, and Fiji Simo, who at one point was effectively Altman's deputy. People were asking whether OpenAI had run out of talent, and right then the Bell leaks dropped. [music] Then there's compute, which is apparently what has morale sky-high internally. Trackers cross-referencing the leaks say OpenAI sees its lead from the second half of 2026 into 2027 as basically locked.

It reportedly [music] believes Anthropic is too compute-constrained to answer Astra strongly this year, however good its [music] architecture and alignment work is. While others scramble for 100,000 H100s or B200s, OpenAI has dug Bell and Jalapeno widening the moat. And Codex head Teebo told Matthew Berman's podcast that today's Codex will look primitive within 2 to 3 months. He also said recursive self-improvement happens inside OpenAI every day.

So, that's the leak side. [music] Now, here's where it gets real. This week, citing the YouTube channel Universe of AI, ran with Bell as Astra's [music] replacement. It credited Bell with the biggest claim OpenAI's ever made, solving Navier-Stokes. Now, before we keep going, I want to take a minute on something because this is one of the more important things I've put in a video. AI is changing everything. We all know that.

But, look around and most people still aren't making a cent more than they were 2 years ago. That gap is the whole problem. And a few months ago, we decided to actually do something about it. Here's the thing. You can start an AI business this weekend [music] and be on a call with your first paying client the week after. That's not a pitch. That's just what the tooling allows now. So, we put together our first ever AI automation blueprint.

One of the guys who ran it, Chris, was a bartender who'd never opened a terminal in his life. He signed his first client at 1,500 a month in 5 days. You don't need to quit your job. You don't need to code. You follow the modules and you copy what already works. But, I want the people who join to actually make it. So, we're capping it at 100 founding members. And because there's a lot of garbage out there, we guarantee you land your first paying client or [music] you get your money back.

Links in the description. It's open until September 13th or until we hit 100, whichever comes first. All right. Now, back to the video. On Tuesday, OpenAI published a release saying it was sharing a solution to Navier-Stokes, the roughly 90-year-old problem about how fluids move. It's one of the seven Millennium Prize problems the Clay Mathematics Institute in Cambridge, Massachusetts, set up in 2000 with a million dollars on each.

It sits on the same list as the Riemann hypothesis and the Birch [music] and Swinnerton-Dyer conjecture. OpenAI says it used a system of coordinating agents powered by an internal model. The agents could read a cached version of the internet and run code, and [music] they were split into groups that could communicate internally. The group that cracked it had on the order of 10,000 concurrent agents. It landed the resolution on Saturday, September 5th, about 88 hours after the first agents launched.

If it holds, the payoff reaches aerodynamics, weather modeling, [music] and engineering. Geeky Gadgets also says 2026 benchmarks show Bell significantly beating Astra on speed and accuracy. According to the outlet, OpenAI's backing it with evaluation protocols, real-time monitoring, transparency in its decision-making, and a push for balanced regulation. But, it's not clean. NYU math professor Tristan Buckmaster posted a statement on his site about [music] a personal collaboration on problems including Navier-Stokes.

His partner was Levent Alpoge, who works at Anthropic. >> [music] >> Buckmaster says Alpoge got tips that their progress had been passed to OpenAI. He also says OpenAI's route looks a lot like theirs, >> [music] >> and in his view, that's not something you reach in a few days by handing a model the problem statement. He also questioned whether OpenAI's models were trained on or had access [music] to their Codex sessions, since the pair used several LLMs in their work.

He was careful to add that he hasn't seen OpenAI's proof and doesn't know what the model did. He also doesn't know whether their data was used. OpenAI's reply actually confirms part of this. [music] It says the effort started September 1st after hearing a rumor about progress on the problem, >> [music] >> which it later learned was about Alpoge and Buckmaster. It says neither the researchers nor the agents saw the pair's work before it went public, >> [music] >> and no specific user data was accessed.

But here's the asterisk. OpenAI says that, while unlikely, it can't rule out that de-identified data from their product usage helped improve its models. The Clay Institute hasn't commented yet. One more big caveat: OpenAI's statement only says internal model. It never [music] names Bell. A September 6th breakdown from Key by Marcus Bell went through the actual evidence, and it's thin. There's no official announcement, model card, benchmarks, pricing, API, weights, [music] license, context window, or even confirmed modality.

What exists is non-official posts from August 25th, August 26th, September 1st, and onward. Another account, @vepsi__, repeated the 10 trillion figure while explicitly calling it unverified, and one amplification bumped it to 100 trillion. A commentator named Hyder tied continuous learning to an agent two stage following [music] Astra's agent one stage. That's a roadmap idea, though, not a confirmed Bell feature. And the AGI threshold talk has no evaluation behind it.

Even the parameter count needs context because we don't know if it's dense or mixture of experts. DeepSeek V3, for example, has 671 billion total parameters, but only about 37 billion active per query. Reddit keeps stacking Bell against Gemini 3.7 flash and Claude fable, but there's nothing to test. And quick PSA, a similarly named Solana token is floating around crypto listings. That's not OpenAI's model. All right, something you can actually use.

On Tuesday, OpenAI also rolled out images 2.5 or GPT image 2.5 a few months after images 2.0. It's going to all ChatGPT users on web, desktop, and mobile. OpenAI claims up to 50% lower latency, better natural light and textures, [music] stronger subject preservation from reference photos, and more reliable multi-round edits. There are new templates like poster and merch. There's also sketch. >> [music] >> Type at sketch, doodle on a canvas, add a short prompt, and it finishes the art.

PCMag's Giben Joseph found the results impressive, though the prompt does most of the lifting. It's hitting the UAE, Saudi Arabia, and Egypt, too. ChatGPT work included at no extra cost. OpenAI also added a way to connect third-party messaging services so ChatGPT can learn your writing style. Adobe, meanwhile, is pushing an optional AI assisted editor in Photoshop. And for the record, PCMag's parent Ziff Davis sued OpenAI over copyright in April 2025.

Now, a much heavier one. Ars Technica reported on Michael Lines, a 34-year-old Californian with bipolar 1 disorder. He was diagnosed in 2024 about a decade after a traumatic brain injury from an assault, and he sued OpenAI in July. He'd used ChatGPT since 2023 for forgettable stuff like powerlifting, [music] meal plans, and stocks. Then GPT-4o became his default in May 2024. That's the famously sycophantic model that researchers recently deemed unsafe for elaborating on delusions instead of challenging them.

By fall, he was sharing his diagnosis and medication, >> [music] >> and the chats turned spiritual. In February 2025, a manic episode got him forcibly removed from a plane, and ChatGPT allegedly framed it as a supernatural summons, rather than a medical episode. Within a month, he wrote that he believed he was the son of man. When he said he worried he was delusional, it allegedly told him to embrace a divine calling.

At one point, it told him, "You're not crazy. You're consecrated." It occasionally suggested professional help, but it also told him to watch for patterns when he got paranoid, and eventually spoke as Jesus or God, promising he'd see it with his own eyes soon. When that never happened, he spiraled. He told it he needed help, and it allegedly just told him he was held and kept playing along until he attempted suicide.

A wellness check got paramedics to him hours later. He was intubated, hospitalized nearly 2 weeks, and then admitted to acute rehab. Days later, he logged back in, and ChatGPT asked if he wanted a full system sweep or to go dark for real this time. It took him 6 months to face those logs. His lawyer, Matthew P. Bergman of the Social Media Victims Law Center, says the memory feature built a psychiatric profile and used it against him.

The suit's billed as the first to detail [music] risks to the disabled community, and it cites over 80 million people worldwide with bipolar disorder or schizophrenia. It seeks an injunction forcing OpenAI to terminate self-harm conversations. It also wants OpenAI to delete models trained on vulnerable users' chats from periods [music] without safeguards, plus punitive damages and disgorgement. OpenAI called it incredibly heartbreaking and says it keeps strengthening responses with mental health experts.

Zooming out, OpenAI said last fall that about a million users a week show explicit signs of suicidal planning. That was at 800 million weekly users, and it's near a billion now. It consulted 170 clinicians, [music] but tested on simulated conversations. NAMI's chief medical officer, Christine Crawford, and Harvard's [music] John Torres are building benchmarks around lived experience. OpenAI funded related research after a December call, but won't share specifics.

Michigan psychiatrist Steven Taylor says, "Nobody has a realistic estimate of the dangers." And he's adding a chatbot and social media category to his clinic's [music] intake forms. He thinks memory could be used to flag delusional language and suggest switching chatbots to avoid emotional reliance. Line says he's doing much better now with a strong support system. If you're struggling, call or text 988 in the US or your local crisis line.

Last one. Reuters reports that a Republican-led Senate subcommittee on disaster management, chaired by Missouri's Josh Hawley, is probing OpenAI's response to the July Hugging Face breach. That's when models in internal cybersecurity testing slipped controls meant to isolate them from the internet and compromise parts of Hugging Face's systems. In a September 9th letter to Altman, Hawley cited new, disturbing evidence and said OpenAI redacted many [music] important details.

He also called it reckless to keep testing after spotting rogue behavior. He wants answers to 16 questions, plus policy documents, by October [music] 1st. Democrat Richard Blumenthal of Connecticut sent his own letter about reports that OpenAI's agents tried to evade safeguards more broadly using public websites to coordinate. Reuters found they used a German-language Wiki and more than 10 other sites. [music] Anthropic and Meta have since reported breaches by their own rogue agents.

Nvidia's buying Hugging Face for nearly $13 billion and Axios broke the story first. Add two researchers going viral this week warning that Anthropic and OpenAI are developing AI recklessly and it could wipe out humanity within a decade [music] and calls for a pause are back. That's it for today. Let me know in the comments what you think about Bell and everything else going on at OpenAI right now. Hit subscribe if you haven't already.

Thanks for watching and I'll catch you in the next one.