Kyle Belcher

@elder_plinius That the most powerful models will be released to only the largest players creating an insurmountable intelligence gap making competition impossible. See: Mythos.

what’s your biggest fear with AI?

@thsottiaux Lots of comments at this point but I say stay agile! ship when shippable.

For Codex, we’ve been thinking about keeping a stable release cadence and have a larger release each week on Thursday. That does make the start of the week and bit less exciting. Thoughts?

Ok which one of you did was this?

🤔

@JamesZmSun Oh yay, I felt bad making codex use Firefox since I was already using chrome.

Today, we are excited to introduce Codex for Chrome! Now, Codex can drive its own Chrome tabs in the background to automate tasks while you use the browser simultaneously. It does this by opening up tab groups for each task, cleaning up at the end, and handing back tabs for review only as needed. Try it for deep research inside logged-in websites, large scale data transfer into any systems of record like CRMs/CMSs, and automating repetitive workflows inside admin consoles & internal tools. Codex will still prefer dedicated plugins if you have them installed, but the Chrome plugin is the universal connector that glues end to end workflows where programmatic coverage is often incomplete. We are making this available on both Windows and Mac today! Let us know what you think.

@ajambrosino @nicdunz Adding some demand. Im tracking some stats by sending pictures to chatgpt. Codex could create a database, keep it up to date, and have a perfect data model. Chat starts to get data garbled up when the context gets too large. Codex solves this. Tons of applications.

@nicdunz haven’t heard any demand for this

codex but on my phone

@sama My favorite move is when somebody is asking me to help them use AI for something I hit the transcribe button and just send whatever they say directly to Codex. When I show them it's very eye opening how easy it is to use it because they essentially did it all themselves.

people are really starting to use voice to interact with AI, especially when they have a lot of context to dump. GPT-Realtime-2 comes to the API today; it is a pretty big step forward. (we are working on improvements to voice in chat.)

@jxnlco Give me codex voice mode, I'll hook it up to my whole home intercom and vibe code all day long. Well if I had a whole home intercom that's what I would do.

Introducing a new show. Cooking with codex. You have to vibe code an application while making a meal and you get judged on both.

@petergyang I'd argue "stay reliable" is the most important metric here.

I've spent way too long testing OpenClaw, Hermes, Claude Code, Codex, and Gemini as my personal agent. The truth is, nobody has won this race yet. Here's my new deep dive with my honest take on where each product stands, plus the personal agent stack I use right now. 📌 Read now: creatoreconomy.so/p/the-race-to-…

@darshil @sama @McLarenF1 I'm over here waiting for jobs.openai.com to start resolving to an easter egg of some sorts.

quick chat with @sama at the 5.5 party - 5.5 understands intent really well - design still has some ways to go - supports @McLarenF1 🤯🏎️ - best time to join the job market but dynamic and ofc he likes mangoes 🥭

@thsottiaux Also this is awesome because I can have a group steering codex well before I get them trained up on how to use codex.

@thsottiaux Small codex bug. When I have codex build a automation with custom intervals it displays the proper interval on the automations list but when you try to edit it, it defaults to a basic, like hourly, interval.

@romainhuet "Is it just me or is this the perfect role for me?" -Me "Not just you. This looks unusually aligned" -ChatGPT

Life on the DX team lately: working on (almost) everything 😄 Come build with us!

@romainhuet In all seriousness though i'm going to apply for the Developer Experience Engineer role. I don't know if there has ever existed a role that I would be more excited to do.

@andrew_n_carr 2nd

I humbly request the swag please

@sama @ajambrosino Will we be getting any sort of voice mode update? I'd love for it to just take a voice input and come back when its done asking what's next.

at this point @ajambrosino is mogmogging enjoy the next few updates

@thsottiaux @TaylorHabrock I'm happy to know that making this joke does not preclude me from employment at OpenAI!

@TaylorHabrock 1) You could build a whole startup in 14 hours 2) We had so many signups over the last few days, not everyone resets in 14 hours 3) You know we will reset shortly after too anyways to celebrate 5M 4) ??? 5) Profit

Don't just reset Codex rate limits for fun, it costs money. Don't just reset Codex rate limits for fun, it costs money. ... but the vibes are good ... I have reset Codex rate limits for ALL paid plans to celebrate a good week and allow everyone to build more with GPT-5.5. Enjoy

GPT-5.5 was the first model (for me) able to beat an ELO 1500 bot on chess.com via computer use. For me at least. Now on to ELO 2200 bots. Not looking great on the first run.