Vasant Dhar

Yeah, you're right.

You know, when I went to Wall Street and I brought machine learning to Wall Street, it was AI was sort of, you know, it was a thing that people viewed with suspicion.

You know, you had economists who were used to sort of linear models and you had physicists who were used to, you know, physical kinds of models and machine learning sort of fell in this awkward space.

They didn't quite know what to make of it.

My objective really was to see if I could get a machine to learn how to trade.

That was just an open question.

It wasn't clear to me that that was even possible.

You know, after a few years, I had a conversation with Scott, actually, in 2015, you know, called, Should You Trust Your Money to a Robot?

And I sort of broke the investment landscape into three holding periods, you know, high-frequency, short-term, and long-term.

High-frequency being intraday, short-term being days to weeks, and long-term being like weeks to months to years.

And at that time, my position was that AI had tremendous potential in investing in the high-frequency and the short-term space because there was sufficient data and machine learning methods were actually capable of picking up on the nuances in markets that the cruder linear models sort of missed.

And I recall that Scott ended that conversation by saying, okay, so trading flows will disappear, but private equity and long-term investing is safe.

And I said, yeah, that's pretty much the case.

And ironically, I'd had a similar conversation with Damodaran in 2015 about whether we could simulate him.

And at that time, we felt it just wasn't possible with the tools.

But in 2022, with the emergence of LLMs, we sort of revisited that.

And now three years later, three plus years later, we're getting ready to release the Damodaran bot next month for commercial use.

And

I'm fairly confident that we've actually managed to simulate his kind of thinking to a reasonable degree.

So to the extent that you actually believe in sort of a free cash flow model to the firm, you know, fundamentals approach to investing, we've actually managed to do that reasonably well.