Peter Welinder
2.2K posts
Peter Welinder
@npew
VP and GM @OpenAI
San Francisco, CA Katılım Nisan 2010
940 Takip Edilen43.5K Takipçiler
@npew Sorry to bother you, there are too many bugs. Here is the first one: x.com/babyghostuk/st…
Peter Leung@babyghostuk
Enabled the Remote Control with my iPad, then my Codex Desktop is now unable to load the plugin page at all..... So frustrating. None of the OpenAI people actually care about these bug reports. I am so sick of it.
English
Big unlock. This is how I've been using Codex for the past few weeks. Excited it's out for everyone.
OpenAI@OpenAI
You've been asking for this one... Now in preview: Codex in the ChatGPT mobile app. Start new work, review outputs, steer execution, and approve next steps, all from the ChatGPT mobile app. Codex will keep running on your laptop, Mac mini, or devbox.
English
A few things I want:
- Faster / smarter loading of the thread. Stalls the whole app while loading. Seems like the latest messages could load first?
- Faster switching between threads across hosts. Not sure what the best UI is, but it takes many taps to go between threads. Even worse when I have multiple hosts.
- Some way to dismiss a "completed task" notification without opening it.
English
Codex on iPad is so awesome. Managing sessions on both my Mac Mini and VPS. Coding on the iPad is finally here.
English
Peter Welinder retweetledi
Codex Thursdays, Fifth Edition:
- Codex mobile (preview)
- Remote SSH/devboxes
- Hooks
- Access tokens
- HIPAA
English
Peter Welinder retweetledi
Peter Welinder retweetledi
It's so weird how difficult it is for the left to understand that all you have to do to decrease housing costs is to let people build more housing.
M. Nolan Gray 🥑@mnolangray
For 45 years, Berkeley built virtually no new housing. By the mid-2010s, it was the most expensive college town in America. Shortly thereafter, YIMBYs took over and kicked off a building boom. Today, nominal rents are below 2018 rates—remarkable progress on affordability.
English
GPT 5.5 xhigh is a beast. Fun to see progress on this (very hard) benchmark.
Kilian Lieret@KLieret
The first ProgramBench task was just solved by GPT 5.5 high/xhigh. Interestingly, high/xhigh picked two different languages for the task (C vs Python). GPT 5.5 xhigh was significantly better than Opus 4.7 xhigh in all metrics. 🧵
English
Peter Welinder retweetledi
After being a Claude Code devotee for a year, I finally tried Codex on a new project this weekend. Once again, in the matter of a few months, it feels like the world changed.
I can see myself doing *everything* inside of Codex this week.
English
Peter Welinder retweetledi
Inspiring to see what students are building with ChatGPT.
OpenAI@OpenAI
Introducing the ChatGPT Futures Class of 2026—26 honorees from the first graduating class to have had ChatGPT throughout all four years of university, who used AI to: - Map 1.5M previously unknown objects in space - Detect disaster survivors through walls and debris - Make 100M+ galaxy images searchable - Preserve endangered languages - Build infrastructure to reroute 5M+ pounds of unsold inventory from landfills
English
Peter Welinder retweetledi
OpenAI model release: We’re throwing a party 🎉 Everything is scribbles and Pets are in Codex. Hope you like goblins!
Anthropic model release: In research preview, it hacked the full Internet for fun. Also, it’s coming for YOUR job specifically. Enjoy the permanent underclass!
English
Peter Welinder retweetledi
I can't figure out if vaccines work or not.
Tough one.
Need Sherlock Holmes on this one.
English
Peter Welinder retweetledi
you can sign in to openclaw with your chatgpt account now and use your subscription there!
happy lobstering.
English
Peter Welinder retweetledi
OpenAI won the hearts and the wallets back.
And honestly, rightfully so.
GPT-5.5 feels like the first model in a while that is not just impressive on paper, but actually changes how people build, write, code, and think.
The market is voting with usage.
English
Feels like we should uplevel the architecture of data centers to the temples of compute they are.
Mike Bird@Birdyword
Many people do not seem to want data centres built near them, despite the fact that they don't cause that much traffic and often generate a lot of local tax revenue. I suspect it's partly because they're ugly! My proposal:
English
Peter Welinder retweetledi
Peter Welinder retweetledi
GPT-5.5 is on par with Claude Mythos
- GPT-5.5 average pass rate of 71.4% (±8.0%)
- Mythos Preview 68.6% (±8.7%)
- GPT-5.5 solved a task that takes a human expert ~12 hours in under 11 minutes at a cost of $1.73
AI Security Institute@AISecurityInst
OpenAI’s GPT-5.5 is the second model to complete one of our multi-step cyber-attack simulations end-to-end 🧵
English
Speeding up Computer Use in Codex.
Ari Weinstein@AriX
Computer Use runs this use case 42% faster in today's Codex app update.
English