I think that one maybe tension that exists still today and still being figured out, is you have these open source models that are public, free, and unencumbered. You have the APIs that are providing other startups or trying to provide as well where I think we’re trying to behave ...

What does the world look where I can just perfectly communicate with anyone else? Does that sort of break down barriers or make it easier to think more commonly about what’s good for the world? Seems a very exciting sort of direction in time.

That makes a lot of sense. It makes total sense the approach that you’re describing. I feel like low-resource languages in particular seemed a great way of increasing the accessibility of these models broadly. I wanted to also say that I’m really curious what’s going to happen as you get ...

Yeah. Including errors, yeah.

Yeah, exactly. That’s one area. Happy to talk more about exactly the details of the technologies that I’m most excited about. Also I’ll hear from you about what the PC is, the needs and opportunities.

We’re finding that in our own work, internally, it’s very useful to be able to turn ChatGPT or internal ChatGPT on the problem as if it were a coworker or assistant.

I think there’s also a lot service governments can do on top of these technologies. If you have issues with misinformation, around bias, I think actually these models themselves are quite capable and can augment policymakers trying to address these problems directly.

Part of what I’m excited to talk about is just I feel like one, there’s a lot of opportunity for startups. This is not what we’re more familiar with as far as investing in new and promising sorts of businesses. There’s possible opportunities to engage further in Taiwan and to ...

Yeah. One topic we’d love to talk to you about is all the recent progress and opportunity that’s been created by advances in AI. With language models like ChatGPT, we’re finding these crossed some pretty interesting qualitative level into being really useful for a broad range of…

Really excited to come talk to you a little bit about what’s been happening recently in that space, and hear from you about how you think about it as Digital Minister in Taiwan.

Part of that I was actually in a PhD program at Berkeley working on robotics and machine learning with Pieter Abbeel in his lab. I took a brief detour to work in tech, at startups, Dropbox for about four years. I would say my background has always been in AI ...

I work at OpenAI. I’ve been there for about six years or so. I’m doing a lot of work, initially, on reinforcement learning and, nowadays, on optimization large language models, as is the style.

That’s how you know Dave and Rose, right?