Episode summary
From Vibe Coding to Vibe Researching: OpenAI’s Mark Chen and Jakub Pachocki
The a16z Show
· · Hosts: Erik Torenberg, Anjney Midha, Sarah Wang · Guests: Jakub Pachocki, Mark Chen
Summary
OpenAI’s Mark Chen and Jakub Pachocki frame GPT-5 as a step toward making reasoning the default, instead of forcing users to choose between instant-response models and slower, more deliberative ones. They say the goal is to remove that friction while improving performance across difficult domains, especially math, programming, and eventually economically relevant discovery.
A major theme is how OpenAI measures progress now that many traditional benchmarks are close to saturated. The guests argue that raw score gains matter less than whether models can autonomously discover something new, sustain longer chains of thought, and operate reliably over hours rather than minutes. They expect the next phase of evals to emphasize real discovery and longer-horizon autonomy, including an automated researcher that can generate new ideas with economic impact.
They also argue that reinforcement learning remains underappreciated because it combines with language modeling in a flexible way, and that coding models are becoming useful enough to change default workflows. At the organizational level, they stress protecting fundamental research, prioritizing compute carefully, and hiring people with strong technical fundamentals and persistence, not just public visibility.
Key takeaways
- GPT-5 was designed to make reasoning more mainstream by unifying fast answers and deeper deliberation by default.
- The guests said saturated benchmarks matter less than evals that test real discovery, long-horizon autonomy, and economically relevant progress.
- OpenAI’s next big target is an automated researcher that can generate new ideas, not just solve known problems.
- They said reinforcement learning keeps working because it can be paired with language modeling to target specific reasoning domains.
- They argued that strong research culture depends on protecting fundamental research, prioritizing compute, and hiring for persistence plus technical depth.
Notable quotes
“The big thing that we are targeting is producing an automated researcher”
“So we think the future is about reasoning more and more about reasoning more and more about agents”
“the next set of evolves and milestone that we're looking at will involve actual um discovery and and actual um movement on on on things that are that are economically relevant”
“I do think you know the future hopefully will be vibe researching”
“we are in the business of doing fundamental research”
How the episode opens
The big thing that we are targeting is producing an automated researcher. So automating the discovery of new ideas. The next set of evals and milestone that we're looking at will involve actual movement on things that are economically relevant.
And I was talking to some some high schoolers and they're saying, "Oh, you know, actually the default way to code is vibe coding. I I do think you know the future hopefully will be vibe researching. Thanks for coming Jacob and Mark.
Jacob, you're the chief scientist at OpenAI. Mark, you are the chief research officer at OpenAI and you guys have the both the uh the privilege and the stress of running probably one of the most high-profile research teams in AI. And so we're just really stoked um to talk with you about a whole bunch of things we've been curious about, including GPD5, which was, you know, one of the most exciting updates to come out of Open in recent times.
And then stepping back, how you build a research team that can do not just GPD5 but codeex and chat GPT and an API uh business and can weave all of the many different bets you guys have across modalities, across product form factors um into one coherent research culture and story. And so to kick things off, why don't we start with GPD5? Just tell us a little bit about the GPD5 launch from your perspective.
How did it go? So I think GPT5 was really our attempt to bring reasoning into the mainstream and um prior to GPT5 right we have two different series of models you had uh the GPT kind of 2 3 4 series which were kind of these instant response models and then we had an O series which uh essentially thought for a very long time and then gave you the best answer that it could give.
From the episode’s captions: roughly the first two minutes, lightly cleaned. Keep watching on YouTube →
Topics
- gpt-5 evaluation and milestones
- agentic systems stability
- reinforcement learning reward modeling
- automated research
- compute prioritization
Companies and tickers mentioned
- OpenAI
- a16z
Watch the episode
Watch “From Vibe Coding to Vibe Researching: OpenAI’s Mark Chen and Jakub Pachocki” on YouTube
Summary, takeaways and quote selection by Octagon. Quotes are short excerpts; the full conversation is in the episode.
More from The a16z Show
-
Cyberattacks Used to Take Months. AI Can Do It in Minutes. -
How AI Could Bring Halo and Other Classic Games Back to Life -
Mike Volpi on Why AI Infrastructure Will Outlast the Hype Cycle -
Why HubSpot’s Co-Founder Is Betting Against the SaaSpocalypse -
Inside Salesforce’s Acquisition Strategy for the AI Era -
How AI Is Turning Figma’s Finance Team Into Builders
The a16z Show, this episode and its artwork belong to Andreessen Horowitz. Octagon is not affiliated with the show; this page is an independent summary.
