OpenAI DevDay - What Actually Matters for Builders

Author: Sam Witteveen · Watch on YouTube

Video authored by the channel shown, not by IANIA; all rights belong to the creator. Transcript and summary generated automatically; the original-language content is preserved.

Original language: English

Original video: OpenAI DevDay - What Actually Matters for Builders · Channel: Sam Witteveen · 18:42 · Watch on YouTube

Topics: AI agents, AI developer tools, Model inference and optimization, Markets and economy

Summary

Sam Witteveen reviews OpenAI DevDay announcements and assesses their significance for consumers, developers, and organizations. He discusses the new personal agent, developer tools and models, Codex updates, faster inference, and changes to subscription plans. He sees potential in managed agents and sign-in with ChatGPT, while questioning whether OpenAI’s app marketplace and integrations will gain traction. Overall, he frames the event as a useful indicator of how AI companies are pursuing practical products and sustainable business models.

Key points

Dots is presented as an always-on personal agent in ChatGPT and Codex, able to use connected apps and take on ongoing responsibilities.
Talking to a Dot is included in a user’s plan, but work it performs on the user’s behalf counts toward usage.
The Agents API provides a managed runtime, including a harness, hosting, memory, multi-agent controls, and computer use.
The Decisions API uses a smaller Luna model to choose from predefined options and launches with image support.
GPT 6.1 Soul is described as approaching Astra-level performance at a lower price, with cached input especially relevant to agent workloads.
Codex updates include cloud workflows across devices, voice, computer use, and security scanning.
Ultrafast offers Astra at higher speed and higher cost, which may be useful when latency matters.
OpenAI introduced a $500 Pro plan and changed the usage allowance of the $200 plan, raising questions about pricing and customer choices.
Sign-in with ChatGPT could enable apps to use tokens included with a user’s plan, though competition among login options may complicate adoption.

Chapters

Start	Chapter	Summary
00:00	Dots and the personal agent	The video introduces Dots as an always-on agent and considers its proactive role, usage model, and possible appeal to everyday users.
05:08	Agents API and Decisions API	The discussion turns to OpenAI’s managed agent runtime and a smaller model API for selecting among predefined options.
07:46	GPT 6.1 Soul and model pricing	The host compares Soul’s claimed performance and cost with Astra and explains why cached tokens matter for agent workloads.
09:20	Codex updates and Ultrafast	The video covers Codex features across devices, voice and computer use, then assesses the trade-off of paying more for faster inference.
12:28	Pro plans and usage limits	The host examines the new $500 plan, the revised $200 plan allowance, and how pricing may affect customer choices.
14:01	Finding customers and app integrations	The discussion reviews OpenAI’s past app efforts, sign-in with ChatGPT, in-chat app extensions, and the new marketplace.
17:08	What the announcements signal	The host reflects on the potential of managed agents and Dots, and on the wider search for useful AI products and sustainable pricing.
18:10	Closing thoughts	The video concludes that DevDay offers a useful snapshot of OpenAI’s plans for consumers and developers.

Quotes

talking to it is free, but the work that it does for you is not going to be free. — 03:05

for agent workloads, the cash price is going to be the real price in here. — 08:49

where latency is the product. — 11:57

We know that the all you can eat buffet of tokens cannot go on forever as costs go up. — 17:39

Concepts and entities

Dots, Agents API, Decisions API, GPT-6.1 Soul, Astra, Luna, Codex, OpenAI, Anthropic, Sam Witteveen

Full transcript (original language)

00:00 Okay, so open AI had their Dev Day yesterday and there are a number of sort of really interesting points in there. What I want to do in this video is go through what they announced and give you my honest take on some of the different things that are there. I actually think there were some really interesting announcements not only for open AI themselves but also about how people are thinking at the frontier model labs themselves. All right, so first up is dots. So what actually is a dot? It’s basically an always-on agent that lives

00:31 inside your chat GPT or at least chat GPT and Codex for now. There were some really interesting things though where you can sort of text it, you can call it and stuff like that. They talk about that it has its own computer in the cloud, its own browser. It can write and test code for you. It uses the same plugins that you’ve already connected to your account. And all those sort of plugins and apps and connectors, there are thousands of them there. One of the cool things here is you can add it to

01:01 Slack. So in many ways you could kind of think of open AI basically covering off what anyone else has done in the personal agent space. So I think the big thing here is that this is their first or at least their more substantial foray into personal agents. So the idea here of where it can basically do things for you, it can be probably a lot more proactive than just reactive. Now it’s going to be interesting to see how that

01:33 proactiveness actually plays out. Is this just a bunch of sort of cron jobs going on underneath or is there something a bit more to it? Obviously a lot of the ideas for this have come from things like Muse, like Grokbot, like Instinct, which is a startup that is very hot at the moment in San Francisco. It’s just raised about a billion dollars on a $10 billion valuation for their first round. And of course, the other two personal ones here would be Open Claw and Hermes.

02:04 And don’t forget, while OpenAI didn’t buy Open Claw directly, they did hire the creator of that. And you can imagine he’s had a lot of input on where dots is actually going and how that should actually work. So, the interesting thing here is that this seems both be a pitch that this is your personal agent, but also perhaps more than your personal agent. They talk about the you give it responsibility. So, rather than just sort of go and do a task for me, like book a flight or something, this is going to sit on top of things and manage

02:34 and monitor things, whether that’s like incoming bug reports, whether that’s testing fixes, whether that’s doing a whole bunch of different things. Now, the interesting thing is that this is actually powered by Astra. So, it’s powered by the top-level model that they’ve got there. And it does seem, in a very sort of Open Claw / Hermes kind of way, that you’re kind of expected to treat this as your chief of staff. Now, the usage model for this is really kind of interesting and worth understanding as well. OpenAI said that your dot is

03:05 included in your plan. So, talking to it doesn’t consume your token usage sort of stuff. But, anything that it kind of does on your behalf does. So, a dot operates on top of your ChatGPT and your Codex threads, but the threads that it drives for you in those particular apps, so in its own computer, it looks like they’re going to be threads that do count to your token usage. So, basically, talking to it is free, but the work that it does for you is not

03:35 going to be free. I can still see that this is going to be a huge win for people who perhaps have never tried out Open Claw, never tried Hermes. And in many ways, Google tried to do this with Gemini Spark after Google I/O, but it seems to me that that hasn’t taken off as much as they would have liked, perhaps in large degree because of the models that were actually running it and what it actually had access to. You could imagine, too, that the actual dots or the dot that works for you is going

04:06 to be a large part of the OpenAI hardware that supposedly they’re bringing out a product either late this year or next year sometime. And so, if that product is like people have pictured where it’s some kind of, you know, hockey puck or something like that that you can then talk to, that you can then interact with, you could imagine that that is something then that people are going to basically put on their desk, have it sitting there on all the time, being able to talk to their dot. And if they feel the usage for that is free for the actual conversations, but

04:37 they’re just getting billed for anytime this thing goes off and does something for you, I can see that being appealing to a lot of people who don’t have a plus account, don’t have a Teams account, don’t have a pro account currently. And like I’ve said in a number of videos, if you’re watching my videos, you’re definitely one of the earlier adopters. I think some of the real huge growth that we’re going to see over the next couple of years with these particular companies and with AI itself is just the normies or the sort of masses are going to come along and start using these things in their day-to-day life. All

05:08 right, so let’s look at some of the developer stuff that they launched. So, that brings us to sort of agents API. And this is basically their managed agent runtime. So, just like we’ve got Claude managed agents, we’ve got a number of companies that are doing this and I’ve made videos about these specific things. You basically get a harness, hosting, memory, sort of multi-agent controls, and now they’ve added computer use to that as well. I do wonder how much is this a me-too product, basically where they just don’t want to lose market share to Anthropic

05:39 who’s been offering their Claude managed agents since much earlier this year. And And interesting that this does seem to be the same stack that powers Codex and dots. Just like the Claude managed agents is using a lot of the same stack that powers Claude code and co-work. So, the way I think about this one is that there are a lot of companies out there that are using the open AI models, whether that’s directly, whether that’s through AWS or Microsoft. This gives them a way to build vertical harnesses so that an agent for insurance claims,

06:12 an agent for sort of contract reviews, all of those things are going to be possible. And it’s basically going to be running on a runtime that’s kind of been rubber stamped by open AI as to what they’re saying is the best way for running their own agents. The second big developer news here was the whole decisions API. And I kind of feel this one really wasn’t a surprise. I’ve made comments before that probably most of the frontier labs have Jev models in development after seeing just the reactions to the whole idea of a

06:44 decision model. And yet again, like I’ve said in a couple of my videos, it does seem that these people are settling on the name of decision models with open AI actually calling their one the decisions API. The idea here is that this is a Luna model. So, again, you don’t need a huge model to do this. This is something that we’ve looked at in my videos with the open models, etc. Certainly, Luna is actually a bigger size than a lot of the open models that I’ve shown. And pretty much the idea here is very similar to Jev. You’ve got a model, you give it a predefined set of options, and it

07:15 basically sort of picks one out. Interestingly, they are launching with image support. And I kind of feel this is a no-brainer here. You don’t need a big model to do it, and you don’t need a huge amount of training. Open AI didn’t really even hide that. They said on stage, it’s the Luna model. And I think for a lot of organizations that aren’t going to want to add another new company, aren’t going to want to run something like this on prem, this just becomes a no-brainer for them to basically use the open AI version of this. Now, another big developer announcement that they had there was the

07:46 new model. Now, almost every year we get a new model and stuff like that. And at first, my gut reaction was like, “Okay, GPT 6.1 Soul.” Remember, Astra is the really big one. Soul is their version of sort of Opus. My first reaction was like, “Okay, so what? Does this really kind of matter here?” But, it was interesting that just like Opus 5.5 seems to have improved massively on Fable. They’re saying that GPT 6.1 Soul is actually getting up there on par with the Astra level here. And the other

08:18 thing here is that this is like a fifth of the actual Astra costs. So, it’s $2 per million input tokens, $10 per million output tokens, and then amazingly, only 10 cents for caching that input. So, when you compare that to Astra, which is $10 in, $50 out, this is an 80% saving for a model that’s supposed to be on par with that Astra model. Now, it does seem that just like with Fable, there are certain kinds of things where if you’re maxing out the

08:49 thinking on Astra or on Fable, the second tier of models is going to be just slightly not as good there. The other thing that I thought was fascinating here was the actual price for caching. So, it’s a 95% discount. And we know that sort of agents reread the same context over and over again, the system prompt files, the conversation so far. So, for agent workloads, the cash price is going to be the real price in here. And on a personal note, I noticed this yesterday with something that I was working on,

09:20 where I’d done 250 million tokens of using the Opus 5.5 model, yet about 96% of all the tokens were actually already cached. So, the price wasn’t anywhere near what you would think it would be if you were talking about 250 million tokens there. So, out of this, I think if you’re running agents on OpenAI, Soul 6.1 is probably going to become your new default model there. All right, next up that brings us to Codex. And there are a few things that people have been asking here for. So, Codex on Linux, multiple

09:51 folders in one project, Codex on your phone. Sam Altman basically announced those very quickly even before they got to probably some of the interesting things in here. Big one for me here is that whole idea of Codex in the cloud. So, this definitely enables you to basically start a task on your phone, pick it up in your browser, run with it over a lot of different surfaces. And it was kind of funny in Romaine’s demo for that being literally like, “Okay, rewrite the entire back end in Rust.” and then just sort of leaving it and walking away. Interestingly, they’ve

10:23 also brought voice into the app and to the CLI. So, this is their bidirectional audio powered by GPT Live. So, it’s not just doing sort of dictation. Unfortunately, they had some hiccups with that on stage. But, I guess if you’re like myself or most people who are using these agents nowadays, you’re talking to them and not typing to them most of the time anyway. There are also some nice things that they showed around computer use and allowing Codex to actually use that, whether that’s for sort of getting app shots as it’s actually making something for you, and

10:55 some nice stuff around security of where it can basically scan for vulnerabilities and prepare fixes for you to review as it’s actually going through your code base. All right, so next up was one that I think is really kind of interesting and this was ultrafast. So, this is Astra at eight times the speed of the normal API for six times the price. Now, they’re claiming that this can do around 300 tokens per second. I’m guessing this is basically the running on Cerebras version of these models. And as we’re

11:26 all starting to see with running these agents, these agents can take hours. So, I guess if you’ve got the budget to pay for this, being able to get something done at a really fast pace may be worth it. I’m not totally convinced though how much people will actually pay for the the six times the price API. They mentioned that this is out on Astra and coming to Soul very soon. I guess if you were to take the Soul 6.1 at six times the price, it’s only slightly more expensive than the normal speed Astra

11:57 was before. So, like that, maybe it is a good deal. Let me know in the comments. I can certainly see that maybe there are people that are quite happy to pay that premium to get the tokens a lot faster, etc. And it is probably going to make sense where latency is the product. So, for anything that you’re doing sort of live editing or interactive demos or where there’s actually a user sitting there waiting for something to happen, then those tokens being that much faster probably is going to be worth it. And this kind of ties us into one of the

12:28 things that I think is probably controversial that they announced, that there’s a new Pro 500 plan. And as I was watching this, I only kind of twigged what it was as they were sort of rolling out that there were Pro 200, Pro 100 plans as well. So, this is a 500 being $500 per month here. And this is going to have 25 times the usage of Plus. It’s also going to have access to ultra-fast in ChatGPT and Codex. And interestingly,

12:58 that usage is going to carry over into partner apps. So, that’s something that’s going to be interesting to see. Now, they reopened the Pro 200 plan. So, the $200 plan, they actually closed that after Astra for a while. I think they just didn’t want people just hammering it and stuff like that. And the sad thing here is and the thing that perhaps is going to anger a lot of people is that this has gone from being 20 times Plus down to 10 times Plus here. Now, we know that usage limits are something that kind of fluctuates anyway, so it’s

13:29 going to be interesting to see how this actually plays out. And does that drive a lot of people who were paying for the 200 to move to the $500 plan, or does it have the opposite effect and drive more people to go over to Anthropic’s plan? It does seem at this point that the pricing is kind of steering you, and that really now if you’re going to go for the sort of top-tier models, you’re going to have to be paying quite a lot of money to do that. All right, so the third part of the keynote was about helping you to find customers. And here

14:01 basically Sam Altman was pretty candid that they’ve tried different versions of these kind of things before with mixed results. And just to recap those, we’ve seen a bunch of these things. They first started out with ChatGPT plugins back in March ’23. They had GPTs for a while from sort of November ’23. They had the GPT store. They then went and sort of switched off all the original plugins. Around this time last year they announced apps in ChatGPT and the whole sort of apps SDK. And I don’t know about

14:33 for you, but I haven’t really seen apps take off here at all. So it does seem to me that this is something that’s still definitely a work in progress. And I think Sam Altman’s statement of saying that they’d had mixed results with what they’ve launched is in many ways an understatement here. So what have they got new? So I would say the big and most interesting piece here is the whole sign-in with ChatGPT, where users can sign in to your app and use the tokens that come with their ChatGPT plan on your particular app. And I think the

15:05 argument here is that that allows you to basically ship sort of like a free core app. That because you’re no longer having to pay for the inference of tokens, you can sort of give people a taste, and then you can upsell to other parts of it without it costing you a fortune. So I think this really opens up the door to a bunch of new freemium products that we’ll see that use this kind of thing. The challenge is going to be, though, do we very soon see a sign-in with Claude? And then are sites

15:35 going to basically have to allow people to sign in with Claude or to sign in with ChatGPT or to sign in with Gemini? All these ideas are trying to do what the Facebook login button did sort of back 15-plus years ago. But, we don’t just have one major player here. We’ve actually got a few different players. It’s also not clear what are you actually offering if someone comes to your site and they’ve only got a free ChatGPT plugin. Uh you then sort of stopping their inference halfway through a question or halfway through some particular task and

16:06 having to tell them, “Hey, you should go and pay for ChatGPT as opposed to you should pay for my product on my site.” All right, next up a plugin extension. So, these are basically full apps that run inside ChatGPT and Code X. Figma and Adobe both showing versions of their products running in there. And supposedly ChatGPT is going to suggest your plugin in the middle of a conversation. I think we’re still in the same place as last year where we don’t know how it’s going to suggest that I use Adobe Photoshop over some other

16:38 hyper cooler image editor that happens to be in their directory as well. Is that going to come down to whichever developer pays OpenAI the most? I don’t know. I guess we’ll see with time. They also launched an OpenAI marketplace with a bunch of different partners including people like Code Rabbit, Notion, etc. And I thought it was kind of funny that Sam Altman actually made the point that because Base Ten, who’s an inference provider, is one of these partners in the marketplace, you will be able to access open-source models in the

17:08 marketplace. My bet is that’s something that you’re not going to see on Claude anytime soon in any way, shape, or form. So, overall I would say that the event was actually very interesting. There were lots of things showing sort of ideas of where things are going. I do think that things like the agents API is going to be something that a lot of people are going to use in organizations to make their hosted agents, if they’re already an open AI customer. I do think it’s going to be really interesting to see how dots plays out. And 3 months from now, are 10 million people using it

17:39 every day? Or is it something that people have just basically added onto the side? And it could be that dots becomes a new part of co-pilot or of some kind of enterprise apps, which is something that they actually talked about making sort of work versions of these kind of helpers as well. Anyway, let me know in the comments what you thought was interesting there. I know a lot of people are going to bag on open AI and criticize everything that they did. And if you’ve listened to the end here, thank you very much cuz I know this video is probably not going to do very well. But I do feel it is very important to see what they’re thinking

18:10 about how things are going to play out. We know that the all you can eat buffet of tokens cannot go on forever as costs go up. We know that lots of people are trying to find the killer app for AI so that you can charge a premium and not just focus on selling tokens all the time. And I think open AI’s dev day is always interesting to get a sort of state of play of where things actually are both for consumers, but for us as developers as well. So, anyway, as always, if you like the video, please

18:42 click like and subscribe and I will talk to you in the next video. Bye for now.