Invest Like the Best with Patrick O'Shaughnessy
01:02:29 · Eric Vishria
[01:02:29] S2: actually funny to me, some of the people are so, so smart, and yet they're in this tech world where they're like reaching these deterministic, almost conclusions of like mass unemployment and all of these different things. I'm gonna give you a really concrete example, which I think is just so good. Take Jeff Henton in radiology. So I think it was twenty sixteen where he was like, "We should stop training radiologists. " AI's gonna do it all better. Jeff Hinton's three orders of magnitude smarter than I am could not have been more wrong, but the actual thing that led him to make that statement and that conclusion was a hundred percent correct. If you look at these radiology images, we should be able to train AI to do a better job reading these things than humans, and that's probably true. And actually, I think the studies in areas have shown that to be true, and we have an investment in a company called New Lantern, which is approaching this. But the big hurdle and the big thing they articulated was like, "Wait a minute, first off, all of the aggregated training dataset doesn't exist anywhere. What you see is companies going after like chest CTs or like very specific elements, but your typical radiologist looks at a whole variety of things every single day, from X-rays to CTs to MRIs of all parts of the body and everything else. And so an AI climbing in- In specific areas like chest CTs, is very marginally helpful because it's only doing that one thing, which could be one of twenty things or forty scans that they read that day. Problem number one, you don't have the data, just like we talked about in robotics and everything else, to train the AI. Problem number two, the whole healthcare industry is oriented around reimbursing doctors for making readouts. How is that gonna work? And there's liability associated with that, and there's repercussions of getting something wrong or missing something, and there's medical malpractice and everything else. How are we gonna avoid that? And how are we gonna get around that? Problem number two, real world stickiness.
01:04:21 · Eric Vishria
We're gonna end up with this application where AI really does help radiologists, it helps radiologists get more and more higher, higher throughput because the AI can do some parts and the radiologist does some parts and they're checking each other and everything else, and you do kind of weirdly end up in this copilot situation for some time, and then you're gonna slowly have the AI. Read more and more of the scans and build up and build up and build up, but the actual duration to get from here to there is gonna take a long time. In the ensuing time, we need more radiologists, not less, because, oh, by the way, everyone's getting more imaging than they used to get because the cost of imaging is going down in a Jevons paradox kind of way. My point on it is, you have someone very, very smart who really understands the capabilities, really understands what's happening, has the right data, but by not thinking of that data in the real world comes to the wrong conclusion. And that's how I think of the unemployment thing. I think it's just, it's almost the exact same setup. Eric, I loved talking about
01:05:20 · Patrick O'Shaughnessy
[01:05:20] S0: markets and COVID with you. An absolute blast. Thanks for the time. Thank you. If you enjoyed this episode, visit colossus dot com. You'll find every episode of this podcast, complete with hand edited transcripts. You can also subscribe to Colossus, our quarterly print, digital, and private audio publication, featuring in depth profiles of the founders, investors, and companies that we admire most. Learn more at coloss You know how small advantages compound over time? That's true in investing and just as true in how you run your company. Your spending system is your capital allocation system. Strategy. Ramp makes it smarter by default, better data, better decisions, better economics over time. See how at ramp dot com slash invest. As your business grows, Vanta scales with you, automating compliance and giving you a single source of truth for security and risk. Learn more at vanta dot com slash invest. The best AI and software companies from OpenAI to Cursor to Perplexity use WorkOS to become enterprise ready overnight, not in months. Visit workos dot com to skip the unglamorous infrastructure work and focus on your product. Ridgeline is redefining asset Management technology as a true partner, not just a software vendor. They've helped firms five x in scale, enabling faster growth, smarter operations, and a competitive edge. Visit ridgelineapps dot com to see what they can unlock for your firm. Every investment firm is unique, and generic AI doesn't understand your process. Rogo does. It's an AI platform built specifically for Wall Street, connected to your data, understanding your process, and producing real outputs. Check them out at rogo dot ai slash invest.