Inside Deep Research with Isa Fulford: Building the Future of AI Agents

**Sarah** (0:05)
Hi, listeners, and welcome back to No Priors. Today, I'm joined by Isa Fulford, one of the pioneering minds behind OpenAI's deep research. This is a new HNTIC product that OpenAI released in February of this year, which uses reasoning and tools like web browsing to complete multi-step research tasks for you. Today, they're making it free to all US users. Welcome, Isa. Isa, thank you for doing this.

**Isa Fulford** (0:27)
Thank you so much for having me.

**Sarah** (0:28)
You and your team have shipped one of the most exciting AI products of late. I use it a lot, deep research. Where did the idea come from? Tell me the origin story.

**Isa Fulford** (0:39)
Yeah. Around a year ago now, we were very excited about the progress internally on this new reinforcement learning algorithm. We were seeing a lot of progress on math problems and science problems and coding problems. At the same time, I was working with my friend, Yash, who works at OpenAI on a few side projects. We're very interested in agents generally and wondered if we could apply the same algorithm to tasks that are maybe more in line with what the average user would do every day. The two first things we were thinking about were online browsing tasks, because I think in a lot of different professions, people do just have to do a lot of research, synthesize a lot of information, and then come back with a report. And then we're also thinking about software engineering. We kind of have been working on those things. I've been focusing on browsing. So to start, we kind of, with the math and coding problems that people were already training on, those datasets already exist. You know, you can have a math problem with a ground truth answer, and you can train on those. But for browsing, it's kind of more open-ended. You don't really have datasets like that that exist. So we really started by grounding the research and what product use cases we actually wanted. We wanted the final model to be good at. So we literally would write out just a list of things. Like, I hope the model that could find this list of products for me and rank them by these reviews from Reddit or something like that. Or I want it to be able to write a literature review on this topic.

**Sarah** (2:10)
I feel like a lot of people, when they think about, you know, browsing and agents, they land on the same two, three transactional use cases that I actually don't think are particularly inspiring. Right. So it tends to be like order a burger on DoorDash or something like that. Or I feel like ordering flowers is also like a really common one. Why do you think you came up with like such a different set of goals for the agent?

**Isa Fulford** (2:37)
Yeah. So I think before we focused on taking right actions, which those are examples of taking right actions, we wanted to get really good at synthesizing information from a large number of sources and mostly read only tasks. That was for a number of reasons. Firstly, just a huge number of knowledge work professions mostly do that, so it would be quite useful for those groups of people. Secondly, I think the overall goal for OpenAI is to create an AGI that can make new scientific discoveries. And we kind of felt that a prerequisite to that is to be able to synthesize information. If you can't write a literature review, you're not going to be able to write a new scientific paper. So felt very in line with the company broader goals.

**Sarah** (3:18)
It's also very meta, because you have helped make an AGI that makes me better at learning, and it's learning.

**Isa Fulford** (3:25)
Yeah, I hadn't thought of that. I love that. More practically, the read-only task is maybe the safety question is a bit more constrained. So it was a good thing to start with as well.

**Sarah** (3:38)
Yeah, it seems that the read-only space, people were also not nearly as ambitious as you were going in, or you and Yash were going in about. Maybe it could understand this set of things for me. Okay, so you thought of these end evals and come up with a set of tasks that could be autogradable or fit a set of characteristics that made them better fit the algorithms, and then what?

**Isa Fulford** (4:00)
That was actually a huge process in itself. I think we initially had built a demo to pitch people on this idea, and it was no model training involved. It was fully just prompted models with the UI, pitching the vision of what this product could look like.