Speaker 1: is quickly growing in usage and in function, and the new Trump administration is taking a new approach to the powerful technology. President Trump signing an executive order on artificial intelligence that will revoke past government policies his order says act as barriers to American AI innovation. Our next guest is a professor of engineering and law at UCLA, as well as the founder of the UCLA Institute for Technology Law and Policy, Professor John Villasenor. Professor, thanks for taking the time. We appreciate you speaking with us. Thank you very much. Trump reversed Biden's executive order that placed some guardrails on the evolving technology. How important do you think that is to the industry?

Speaker 2: Well, I think generally we're gonna see the Trump administration taking a much more hands-off approach to AI regulation than we saw in the Biden administration. And I think his decision to revoke that October 2023 executive order from the Biden administration is part of that process.

Speaker 1: What were some of the guardrails? I'm sorry? What were some of the guardrails?

Speaker 2: Well, there were some obligations with respect to reporting. For example, the security protections that were used to protect model weights and things like that to the government. And so that's now been revoked. And I think the new administration is, I would expect taking a kind of step back and probably would take more of a clean slate approach to figuring out how and in what manner they want to approach AI regulation.

Speaker 1: Does it concern you at all that people outside of the administration are advising the president on AI policy that are also the ones who might gain the most from it financially?

Speaker 2: Well, I think any administration is going to seek input from people both inside the administration as well as people who are working in the areas that are potentially affected by the regulation. And as long as the administration or any administration takes a balanced view of all the input, I think there's a really good opportunity to create good policy. So I think the fact that the administration is seeking input from a broad range of sources isn't inherently problematic.

Speaker 1: Professor, knowing what you know about the industry, not asking you to take a political stance here, but is one administration's way of dealing with AI better than the other, or is a combination of both the best? What do you think?

Speaker 2: I'm actually more optimistic about AI under this administration than I was under the Biden administration. And the reason is the Biden administration had a very sort of fear-based approach to AI regulation, whereas I think the Trump administration has a more, I think, properly balanced approach in terms of understanding the incredible benefits, but also, of course, being aware that like with any technology, there are going to be some downsides.

Speaker 1: And lastly, where do you think AI and large language model technology will be, say, in the next three to five years? What do you think we can expect?

Speaker 2: You know, I think we can expect the unexpected. I don't think five years ago, any of us expected to see the extraordinary capabilities that we're seeing today that are now in the hands of anybody who wants it, at least in terms of large language models. And so if we've learned anything, it's that predicting the future is really hard, especially when it comes to AI.

Speaker 1: Maybe AI will be able to predict the future better than us at some point.

Speaker 2: We'll see about that. But it's going to be, this administration is coming in at an absolutely foundational time, incredible importance for AI. So there's going to be a lot of very important developments in the next couple of years.

Speaker 1: All right, we know you'll be following it. We will too. Professor John B. Esenior, thanks so much. Thank you so much.