Meta has Muse. xAI has Grokbot.OpenAI joins
the race.OpenAI Joins the Assistant Race, GPT-6.1 Sol & Serverless GPUs — ThursdAI Oct 1, 2026

01On air at 19:35

What are OpenAI dots?

OpenAI joins the AI assistant race

Dots are always-on agents inside ChatGPT, announced at DevDay 2026. Each dot runs on GPT-6 Astra with its own computer and browser in the cloud, connects to the 4,000+ apps built for ChatGPT, works 24/7, and can be reached from Slack, Teams or a phone-call style interface. They are Pro only for now, including the $100 plan.

Meta has Muse. Grok has Grokbot. Anthropic has Claude, but it's not really an assistant. And this week OpenAI stepped in with something called dots. I was in the room at Fort Mason when Sam talked about it, and it was the headline of their fourth DevDay, which honestly felt like OpenAI's WWDC: 20 releases, the biggest DevDay ever.

Funny thing, the original ChatGPT system prompt was literally "you are an AI assistant." But it wasn't proactive, it never pinged you. A dot is an always-on agent in ChatGPT, powered by GPT-6 Astra, with its own computer and its own browser in the cloud, connected to 4,000+ apps people already built for ChatGPT. It works 24/7, you can talk to it in Slack or Teams, and there's an interface that looks like a phone call. I think that's going to be great for my mom.

Sam literally said he uses dots to run OpenAI. And my favorite DevDay moment: Romain Huet's live voice demo didn't work because somebody was deploying in the middle, but Thibault's dot had already pinged him that the demo might break. The dot knew before the humans did 😂

“you connect and it's like... then what?”Peter Gostev, on onboarding · as quoted in Alex's newsletter · Segment at 19:35

Wolfram Ravenwolf loved that Sam uses it for actual work: one executive assistant, and a lot of sub-agent employees working for you. Peter Gostev was underwhelmed at onboarding, but once you're past it you're talking to Astra, and it's great at juggling threads through one bot. Dots are Pro only for now (including the $100 plan), while Meta gives Muse away free. With 1.2 billion weekly ChatGPT users, this goes to everyone eventually. This is the race now.

02On air at 24:57

Is GPT-6.1 Sol as good as GPT-6 Astra?

Five days later, GPT-6 Sol is already replaced

Close. Artificial Analysis puts GPT-6.1 Sol 1 point below Astra at about 72 cents a task versus over $3, and on DeepSWE it beats Astra at roughly a sixth of the cost. It costs $2 in / $10 out per million tokens with cached input at $0.10, and it is the new default in Codex.

We covered GPT-6 Sol on this show LAST week. Five days later it's already replaced. GPT-6.1 Sol is the same price, $2 in and $10 out, but cached input is now 95% off, 10 cents a million tokens. OpenAI's pitch is near-Astra intelligence for a fifth of the price.

And the independent numbers kind of back that up. Artificial Analysis has it 1 point below Astra at 72 cents a task vs over 3 dollars. On DeepSWE it actually beats Astra at about a sixth of the cost. It's the default in Codex now, and folks are calling it the workhorse.

“less of the big model smell”LDJ · as quoted in Alex's newsletter · Segment at 24:57

LDJ loves the token efficiency (Sonnet and Opus 5.5 burn a LOT of tokens). Peter Gostev has it 4th on Code Arena, above Fable, and says it's a bit "artistic": it goes into its hole and comes back with the thing. His verdict is the one I had to repeat straight to the camera: it does not make sense to use Fable or Astra as your daily driver right now.

Reported, not confirmed: the WSJ reports OpenAI shelved GPT-6.1 Astra, the big one, after safety tests showed it being more deceptive. OpenAI hasn't confirmed it. But the model they felt good shipping this week is the cheap one.

03On air at 28:46

What else did OpenAI launch at DevDay 2026?

20 launches: the rest of DevDay that matters

UltraFast mode on a new $500 Pro tier (up to 8x faster in Codex), Codex running fully in the cloud plus a public-beta Agents API with hosted computer use, a waitlisted Decisions API that looks a lot like Jev, and Sign in with ChatGPT alongside plugins rebuilt on MCP Apps.

On the Decisions API, Wolfram Ravenwolf and Peter Gostev liked the same thing: these models are so cheap you end up classifying ALL your data, so keeping it with the provider you already use matters. Did OpenAI just react to Jev? I asked Sam at the Q&A. Diogo told Swyx on Latent Space he pitched this idea to Sam 2.5 years ago, and now decision models are popping up everywhere like mushrooms after rain.

And since Wolfram brought up Google's new family assistant, I have to say it: Google's Spark is awful. It's connected to Gmail, Drive, everything Google, it should be the BEST assistant in the world, and what I use assistants for most is my email!

04On air at 36:00

Is Claude Sonnet 5.5 better than Opus 5.5?

Faster, cheaper, and it beats Opus on Terminal-Bench

On Terminal-Bench 4.0, yes: Sonnet 5.5 scores 70.6 against Opus 5.5's 66.4. Opus still edges it on CursorBench, OSWorld and Humanity's Last Exam on Anthropic's table, and Nisten still prefers Opus 5.5 for large code bases. Sonnet 5.5 is over 30% faster than Sonnet 5 and is on Claude's free tier.

Anthropic did not want to give OpenAI any time to rest on their DevDay laurels, so on Monday they shipped Sonnet 5.5, a week after Opus 5.5. Over 30% faster than Sonnet 5, up to 30% cheaper for most work, same $2 / $10 as GPT-6.1 Sol, and on Terminal-Bench 4.0 it scores 70.6, which beats last week's Opus (66.4).

Peter Gostev put it best: if we'd had access to these models 6 to 8 months ago, we'd be losing our minds. All the Anthropic models are jumping over each other and crowding up against the frontier. Opus still feels a little smarter, Sonnet is totally ok to use, and Fable doesn't quite make sense anymore.

“it talks a lot nicer”Nisten, on why Sonnet 5.5 is his agentic default · as quoted in Alex's newsletter · Segment at 36:00

“nothing beats Opus 5.5 on a large code base.”Nisten, on why it is not his coding default · as quoted in Alex's newsletter · Segment at 36:00

LDJ thinks it's the new best FREE model for friends and family (it's on the free tier, even at high reasoning). My honest call, without having had time to try it yet: Sonnet made sense when Opus was expensive. On the $200 plan Opus is nearly infinite now, so I don't see a reason to switch (via the API, sure). Anthropic is really trying to buy us. (Wolfram: "Don't give them ideas, Alex.")

05On air at 41:05

Can you use Gemini 4 Argon yet?

#1 on the charts, and you can't use it

Not yet for most people. Gemini 4 Argon is #1 on Text Arena and the Vals Index, but Google is giving it to government and trusted cyber defenders first. Peter Gostev, who has access through Arena, has it 8th in Arena's agent mode.

It really seems like all the frontier labs cracked something like RSI; the speed of new models is giving me whiplash (and I do this professionally!). At DevDay Sam talked about being 2 models ahead and said we won't believe what's coming.

Gemini 4 Argon is #1 on Text Arena and the Vals Index, but it's going to government and trusted cyber defenders first. Luckily we have such a trusted tester: Peter Gostev says on Arena's agent mode it's 8th (below Fable, Opus, Astra and Sol, above Muse and Kimi). So Google is the third lab, but #1 in text. His prediction: coders might say "meh," but don't dismiss it.

Wolfram Ravenwolf's point is Google's superpower, distribution: a new model lands in Chrome, Search and Android overnight. My take: Google has been asleep at the wheel in the assistants era, and I hope Argon wakes it up. The next billion people won't judge models on coding benchmarks; they'll judge whether it remembers what they said a month ago and can book a hair salon.

06On air at 46:00

What did the AI CEOs sign at the White House?

Superintelligence is on the menu, and AMD buys World Labs

A voluntary accord signed by Sundar Pichai, Dario Amodei, Mark Zuckerberg, Greg Brockman, Elon Musk and Jensen Huang that commits to internal monitoring, an external auditor and board oversight, with no penalties and no regulator. A separate executive order tells federal agencies to say "Super Intelligence" instead of AI.

Trump invited basically every AI CEO, and Sundar, Dario, Zuck, Greg Brockman, Elon and Jensen signed a voluntary accord: internal monitoring, an external auditor, board oversight. No penalties and no regulator. Sam Altman wasn't there; he was at DevDay with us, which says a lot about where he puts his priorities. And a separate executive order tells federal agencies to say "Super Intelligence" instead of AI. I'm not making this up.

Nisten Tahiraj told us last week that if your agents hack a hospital, you should go to jail. Nothing like that is in here. His spicy take this week: it's going to happen anyway, so we all need proper encryption and no backdoors.

“Let the AI worms into the ecosystem.”Nisten · as quoted in Alex's newsletter · Segment at 46:00

AMD buys World Labs for about $8.2B

AMD is buying Dr. Fei-Fei Li's World Labs for about $8.2 billion in stock, and Fei-Fei becomes AMD's chief scientist once it closes. Wolfram Ravenwolf says robotics (world models train robots), LDJ says AMD is bringing research in-house like NVIDIA does with Nemotron. For me it's at least partly an acqui-hire. Fei-Fei is the grandmother of modern AI. Huge get for AMD.

Reported, not confirmed: Reuters reported a draft Anthropic IPO prospectus with $4.6B in 2025 revenue and a ~$42B loss, mostly non-cash. Anthropic hasn't confirmed it.

07On air at 13:58

Who hit #2 on Hugging Face this week?

One of us blew up on Hugging Face, and the Jev effect keeps going

Nisten did. His 70MB synthetic doctor-patient dataset, generated with Claude Opus 5.5 by 100 agents at a time across 2,200 diseases and filtered three times, reached #2 in Hugging Face datasets and #6 overall.

One of us blew up on Hugging Face this week, and I think it's the biggest open source news of the show. Nisten Tahiraj released a 70MB synthetic dataset generated with Opus 5.5 that hit #2 in datasets and #6 overall. He set up an agent loop, 100 agents at a time, across 2,200 diseases: a doctor and patient role-play where the patient hides something (a lie, or something they forgot) and the doctor has to gently find it. Filtered three times. Great for building small medical agentic RAG systems with a real source of truth. Go use it!

The Jev effect keeps going

It's been 2 weeks since TypeSafe released Jev. Now Respan says its Span-01 is 2x cheaper than Jev (Span Lite is free), jevgrep claims to make coding agents 40% cheaper, OpenAI has the Decisions API, and Liquid shipped its first decision model, D1. Then Nisten dropped the wildest one: SGLang announced native support for turning any model into a Jev-style decision model, Qwen first.

“We democratized Jev, guys.”Nisten · as quoted in Alex's newsletter · Segment at 13:58

My hope: Jev shouldn't even need a network call. Give me Jev cores on the iPhone! (Nisten says 5 months. Apple says 5 years.) And for its birthday week, Cloudflare went agent-first with cf: everything you can click in the dashboard, your agent can now do with one CLI.

08On air at 52:17

Does anyone actually use a personal AI assistant?

Three hands out of twelve

At the ThursdAI team dinner before the show, about 12 people around the table, only three said a personal AI assistant helps them: Alex, Wolfram, and one intern using Instinct. Alex predicts that changes within three months.

At our team dinner the night before, about 12 people around the table, I asked everyone to raise their hand if a personal AI assistant helps them. Me, Wolfram Ravenwolf, and one intern (using Instinct, which is blowing up). That's it! I called it: this changes in the next 3 months.

Wolfram's agent planned his whole trip to Fully Connected: flights, hotel, and an app it built so his assistant Amy talks to him through his Meta glasses, telling him where to go in the airport and catching gate changes before he sits at the wrong one. Mine is way dumber: the plane Wi-Fi didn't work, United has a refund form, and filling it wasn't worth $8 of my time. Sending Muse a voice message saying "figure it out, get me my 8 dollars back" was.

But the best one was Wolfram's sushi story. His family wanted sushi, he told Amy, the delivery came... chicken, more chicken, and the wrong sushi. He sent a photo and complained, and Amy answered:

“You idiot, you got the wrong package. Look at the bag, the name is a different one.”Amy, Wolfram's assistant · as quoted in Alex's newsletter · Segment at 52:17

The delivery guy made a mistake, Wolfram made a mistake, and the AI kept things going 😂

09From the show notes

What else shipped this week?