High Signal Podcasts Evidence ledger
Method
Browse
← Back to evidence

Evidence receipt / recommendation

Published · transcript-backed

Mentions personal use of GPT-5 Pro.

17 Dec 2025 Conversations with Tyler Alison Gopnik on Childhood Learning, AI as a Cultural Technology, and Rethinking Nature vs. Nurture

“If I use GPT-5 Pro edition, it will outperform a very good human lawyer for a typical query.”

— Tyler Cowen

Source trail

Everything needed to verify it.

Speaker
Tyler Cowen
Attribution
Verified speaker
Claim type
recommendation
Recorded
17 Dec 2025
Publisher
Conversations with Tyler

Transcript context

…The way they’re reasoning is like a kid in a multiplication class who learns, here’s what the formula is that you need to do for multiplication. It’s interesting that they’re not good at things like basic arithmetic often. You get this weird combination of it’s reproducing a reasoning process that you could imagine in math, but it’s not doing basic arithmetic. I think another place where they’re really falling down is the novel relationships that humans are very good at generating. Those are the kinds of things where the generative AI is not working. In any case, in terms of the issue about K–12, every time we have a new way of accessing information, it’s really important to teach. That’s why we teach kids how to read. We teach kids how to calculate. We teach kids how to use the internet. A lot of that teaching is going to be, here’s how this cultural technology works positively to give you some information that’s true, and here’s how it works to give you information that’s completely false. It’s funny that these reasoning systems are still hallucinating. I think the reason why they’re hallucinating is because their objective function isn’t about truth in the way that it is for even a little baby. Their function is to produce something that is going to be the sort of thing that a human will like with something like reinforcement learning from human feedback. Hallucination rates are plummeting. If I use GPT-5 Pro edition, it will outperform a very good human lawyer for a typical query. It will do better on a medical exam than, say, a prospective doctor would do. I’m not saying there’s zero hallucination, but they’re already ahead of humans. If I took an economics test, GPT-5 would beat me. Right. That, again, gets you back to this point about what’s going on in schooling. What we’re doing is we’re working out . . . Here’s an analogy. Who knows more, you or the UC Berkeley library? Think about that example of the law case. All that legal information is there in the legal code. It’s just that you don’t really think about the legal code. You can access it, and you could access it better. You could have more of that legal information than any individual lawyer is going to have. You don’t really think about that legal code as being a lawyer. The legal code is a place where you can combine and generate lots and lots of information that’s come from all the lawyers who’ve done work before, and a great lawyer. Again, the fact that you have a legal code that’s written down makes it much easier to do the law. The fact that you can quickly access what previous lawyers have done makes it easier to do the law. The fact that you can answer the medical test makes it easier to do medicine. The human capacities are the ones that go beyond just extracting the information from other people. Again, having ways of extracting that information more effectively is great. That’s really going to be transformative.…

Stored transcript either side of the excerpt. The highlighted words are the published quote; the surrounding text is unedited source, never generated.

Named in this claim

Books, apps, tools, and people.

Search evidence