OpenAI, every mention
57 scenes (2026), the whole family · ← back to OpenAI
tap a year for its mentions
every year 2026 anyone Garry Tan 114Harj Taggar 64Sam Altman 37Diana Hu 34Craig Cannon 21Calvin French-Owen 20Jared Friedman 19François Chollet 11Aravind Srinivas 11Raza Habib 10
Verbatim, from the transcripts: the passages where OpenAI comes up
Multi-GPU Kernels, Intelligence per Watt, Heterogeneous Inference, and More | YC Paper Club · Y Combinator
- ▶ 2:18 Francois Chaubard And then, uh, we have Mark coming up next, a former PyTorch maintainer, uh, co-founded, uh, GPU mode with my friend Casey Elward, um, and then is co-founding, uh, core automation with, uh, OpenAI VP research, Jerry Twerk.
- ▶ 30:44 unnamed speaker So you don't need to, you know, pay, pay Claude or pay OpenAI anymore for, for your LMs.
- ▶ 44:59 unnamed speaker Um, I can tell you about, like, what I think are some important open problems that I don't think, like, get enough, nearly enough attention that are very different from, like, just, like, asking codex to go cook.
- ▶ 1:06:09 unnamed speaker So this is a visualization of OpenAI's old hide-and-seek environment.
- ▶ 1:06:53 unnamed speaker So here's an example of some of the logic from that OpenAI hide and seek environment.
Sam Altman: "Never a Better Time to Do a Startup" · Y Combinator
- ▶ 1:37 Sam Altman You know, it's, I think you could either be sad and be like, oh man, like a Codex prompt is a whole startup, or you could be like, I can go start the world's most ambitious, crazy company.
- ▶ 11:45 Garry Tan I guess I want to get into the beginning of OpenAI in that it actually started as YC research looking at this idea, I mean, kind of a crazy idea that, you know, you and, um, a ragtag crew decided we're going to dedicate our lives to this,…
- ▶ 17:38 Sam Altman OpenAI with, I met, like, many, many years in some cases before, uh, starting OpenAI, and I kind of, like, got to know them over the long, like, a very long career journey, and this happens again and again, and so the sooner you can put… 3 times in the scene
- ▶ 25:32 Sam Altman Now the goalposts have moved, and it's easy to say, well, you know, here were the, here's how this happened, and here are the mistakes that OpenAI made, and we did make some big ones, of course, but these systems have gotten incredibly… 5 times in the scene
- ▶ 31:28 Garry Tan It sounds like it would, um, it would take a similar shape to, uh, creating OpenAI in that, like, do you need to be ready for, uh, people to attack you or dismiss you? 5 times in the scene
Jensen Huang: The Mindset That Built NVIDIA · Y Combinator
- ▶ 25:23 Jensen Huang We have, you know, Boris is in the back, and, and we, we've got cloud code autonomously running in sandboxes all over NVIDIA, and that's really fantastic, and some people use codex, some people use cloud code, some people use cursor, some…
How Two French Engineers In New York Built The Company That Monitors The Entire Cloud · Y Combinator
- ▶ 2:17 Olivier Pomel Uh, I mean, I mean, now, you know, look, now there's, there's, uh, you know, DeepMind and OpenAI and Meta.
The Age Of The 40-Year-Old Solo Founder Is Here · Y Combinator
- ▶ 18:52 Bryant Chou And we've used a collection of models, open AIs, ChatGPT images, but then we've created 3500 prompts for web designs that then Ploy takes inspiration from.
- ▶ 29:30 Garry Tan So there is like a competitive dynamic to it, which has already happened for coding tools, for instance, like sort of unconscionable in, uh, 20, 26 to not be using cloud code or codex or cursor, you know, just.
The Most AI-Pilled CEO We Know · Y Combinator
- ▶ 3:56 Pedro Franceschi Uh, and I think electricity was Opus 4.5 and, and sure Opus models and, and, you know, open AI models got, got better and better since then.
Inside YC's AI Playbook · Y Combinator
- ▶ 12:19 Pete Koomen The harnesses that have gotten really popular, right, uh, Claude Code, Codex, Pi, OpenClaw, Hermes, they're all,
- ▶ 44:43 Garry Tan This should be an extension of yourself and what you care about, not what, you know, Meta or Alphabet or even OpenAI or Anthropic care about.
How The Best Companies Defend Against Mediocrity And Rot · Y Combinator
Tokenmaxxing: How Top Builders Use AI To Do The Work Of 400 Engineers · Y Combinator
Recursion Is The Next Scaling Law In AI · Y Combinator
- ▶ 10:11 Francois Chaubard This starts from, like, literally Tagula-Rasa weights, and it can outperform at that time, if we go back, you know, we had, um, O-three, if you remember back, way back when. 2 times in the scene
How to Make Claude Code Your AI Engineering Team · Y Combinator
This Startup Catches Fraud at Scale · Y Combinator
- ▶ 24:17 Karine Malata Um, I think GPT-IV came out during the batch, which was really interesting, and as we were running this pilot for this first customer, sort of OpenAI was coming out with new models in the middle of the pilot, which was changing, one, our…
François Chollet: Why Scaling Alone Isn’t Enough for AGI · Y Combinator
- ▶ 18:51 François Chollet In particular, the OpenAI O-one and then O-three models, which by the way, they were demonstrated by OpenAI on Arc, because it was the one unsaturated reasoning benchmark that was really showing that this model was different. 3 times in the scene
- ▶ 18:51 François Chollet In particular, the OpenAI O-one and then O-three models, which by the way, they were demonstrated by OpenAI on Arc, because it was the one unsaturated reasoning benchmark that was really showing that this model was different.
- ▶ 18:51 François Chollet In particular, the OpenAI O-one and then O-three models, which by the way, they were demonstrated by OpenAI on Arc, because it was the one unsaturated reasoning benchmark that was really showing that this model was different. 2 times in the scene
- ▶ 32:13 Garry Tan Is the match pair to OpenAI writing? 4 times in the scene
- ▶ 32:28 François Chollet OpenAI worked on, uh, on, uh, in particular on, uh, on Dota too, uh, the OpenAI five model, which was, if I recall correctly, so this was like, not just pre-GPT, but also mostly pre-Transformers because they were working with a stack of… 2 times in the scene
- ▶ 41:46 unnamed speaker We've all here heard of the OpenAI founding story, and something that's always struck with me is just like both Sam and Greg say that it was a little odd in the early days because they didn't actually know what to do.
AI Is Unlocking Millions Of New Builders · Y Combinator
- ▶ 20:27 Madhav Jha How do I get an open API key?
- ▶ 33:16 Mukund Jha Like, for example, you know, um, Opus is obviously a workhorse, um, you know, like, um, Codex is really good in back-end debugging.
The Powerful Alternative To Fine-Tuning · Y Combinator
- ▶ 1:50 unnamed speaker And then Anthropic or OpenAI will come along and just eat your lunch in the next model release. 2 times in the scene
- ▶ 10:26 unnamed speaker It sounds like this is a complete different paradigm than RL, because we went through the S-curve of regular pre-training, RL with when OpenAI release O-one, and now this feels like a new one.
OpenClaw Creator: Why 80% Of Apps Will Disappear · Y Combinator
- ▶ 8:57 Peter Steinberger You sent me a text message, and there was no file ending, so I looked at the header, I found its opus, so I used ffmpeg to convert it to wave, and then I wanted to, like, transcribe it, but didn't have visper installed, but then I looked… 2 times in the scene
- ▶ 19:00 Peter Steinberger Like, I, I love codex, because it, it looks through way more files before, before it decides what to, what to change. 3 times in the scene
We're All Addicted To Claude Code · Y Combinator
- ▶ 1:01 Garry Tan He's one of the first people to create codecs at OpenAI.
- ▶ 1:01 Garry Tan He's one of the first people to create codecs at OpenAI.
- ▶ 1:55 Garry Tan Because you're, I mean, right out there at the forefront of it, I mean, Codex pioneered all of the, a lot of the ideas that now, like, everyone still uses, and Codex is still evolving too.
- ▶ 2:07 Calvin French-Owen For brief context, when I was at OpenAI, um, I was working on the Codex web product.
- ▶ 2:07 Calvin French-Owen For brief context, when I was at OpenAI, um, I was working on the Codex web product. 2 times in the scene
- ▶ 4:19 Diana Hu And I think the fascinating thing is because it's on the terminal is the purist form for composable atomic integrations, because if you came from, uh, ID first world, which is where cursor was, and I suppose codex too, this concept of, uh,…
- ▶ 5:31 Garry Tan I mean, I really like how clean a sandbox conceptually is in Codex, but then I just ran into all these crazy issues, like trying to do, you know, run just simple testing, right? 2 times in the scene
- ▶ 6:26 Calvin French-Owen Like thinking about a cursor or a cloud code or a codex CLI, the fact that you can just download it and use it without having to get IT permissions or anything makes a huge difference.
- ▶ 10:10 Calvin French-Owen I'll like clone the repo and then spin up codex or cloud code and be like, Hey, give me a walkthrough of what's going on here.
- ▶ 10:32 Calvin French-Owen For, uh, I think it was O three, like one of the reasoning models.
- ▶ 11:25 Calvin French-Owen If you look at a codex or a cloud code, uh, they actually just use like grep.
- ▶ 13:41 Calvin French-Owen I mean, I can pick on OpenAI slightly in this example. 2 times in the scene
- ▶ 14:28 Calvin French-Owen Um, I use the cursor bug bot has gotten quite good and I actually like codex for code review as well.
- ▶ 17:17 Calvin French-Owen It's actually interesting, the codex approach is kind of the opposite, and they just wrote about this on the OpenAI blog.
- ▶ 17:17 Calvin French-Owen It's actually interesting, the codex approach is kind of the opposite, and they just wrote about this on the OpenAI blog. 4 times in the scene
- ▶ 18:17 Garry Tan Like, are we at 24 hours or 48 hour running jobs on Codex, and that architecture is correct for that world?
- ▶ 18:54 Calvin French-Owen Whereas OpenAI really leans into this idea of just, like, we are going to train the best model and reinforce over time and get it to do longer and longer horizon things in this pursuit of artificial general intelligence.
- ▶ 23:30 Calvin French-Owen I find this so often open AI, like strolling through the code base. 2 times in the scene
- ▶ 32:13 Calvin French-Owen Like, I'll just tell the quad or codex what to do, and then
- ▶ 40:37 Calvin French-Owen Refreshing complex UI state and Opus often will miss it if there's many files, but Codex seems to catch it.
- ▶ 41:42 Garry Tan I guess what's funny is I tried to use Codex just now, uh, for my Rails project, but the thing is like, it's kind of obvious that nobody at OpenAI cares about Rails, which is fine. 6 times in the scene
- ▶ 41:42 Garry Tan I guess what's funny is I tried to use Codex just now, uh, for my Rails project, but the thing is like, it's kind of obvious that nobody at OpenAI cares about Rails, which is fine. 4 times in the scene