Lazard CEO Peter Orszag Talks AI in Banking

SPEAKER_1 (0:02)

Bloomberg Audio Studios, podcasts, radio, news.

Tom Keene (0:07)

Peter Orszag is with his chairman and chief executive officer of Lazard that barely describes his contribution to modern retirement in America. Coming out of LSE a number of years ago, he wandered in and got a job and he said, you know what, when we take a new job, we should automatically be in the 401k of the company. We say, well, that's no big deal. At the time, it was absolutely seismic. He also looks at his economics as well as running the shop at Lazard. Here is a seismic essay from January this year. The risk of higher US inflation in 2026

Research demonstrates that consumer perception of inflation is based on, here's the grocery part, the frequency of purchase of, say, eggs.

Inflation rising above 4% by the end of 2026 is plausible and a most likely scenario. You and Adam Posen nailed it. Where do we go from here?

Peter Orszag (1:06)

Well, I think the important thing about that is not necessarily the kind of victory lap, but instead, why did we say that? And the reason we said that is we are living in a new supply side era where, if you want to understand inflation today, you need to look at the supply shocks that happened a year or two ago.

And the thought, in fact, I just read a piece of research suggesting, don't worry, the next few months, you'll see a tailing off of inflation because these one-time effects will erode. Okay, but our other point is they're going to be continual new supply shocks. So we're living in this era, supply side era 2.0, in which you're going to be hit by these things on an ongoing basis. And so treating them each as a one-off thing is a mistake.

Tom Keene (1:48)

Our Bloomberg Money audience is hit by it each and every moment wrapped around personal finance and retirement and wealth management. Without getting into the mathiness, does there need to be a new set of a higher interest rate, what the fancy people call a higher actuarial assumption? Are we shifting in our interest rate now?

Peter Orszag (2:08)

I definitely think that the interest rate environment is different today than it was three or four years ago. It's unfortunate, by the way, that we didn't take advantage of that moment when we had extended moment, when we had low rates, to extend the maturity of treasuries, which is something that Bob Rubin and Joe Stiglitz and I had called for.

Instead, we have been shortening maturity. That's too bad. But I think we are in a new era in which, frankly, the bond vigilantes are coming back.

Tom Keene (2:34)

We could digress.

Scarlet Fu (2:35)

I know. If we start talking about the bond market, we can go in a different direction. Getting back to that inflation story that you were just talking about, is AI inflationary? I mean, there's the build out, then there's the implementation, and then there's the application. We haven't totally gotten to the mass application yet.

Peter Orszag (2:50)

Yeah. I think on AI, the build out is massive, maybe three, three and a half percent of GDP investment, which is larger than railroads, highways, electrification by a significant margin. That's also putting upward pressure on interest rates. My own view is I think we need to separate the technology and the business model here. I'm a huge believer in the technology. I use it, probably have 70 or 80 agents running as we speak.

I'm astonished by how rapidly things are progressing. But because so much of the investment is debt financed, I do think the risk of financial distress in this ecosystem is higher than people appreciate. Because when it's debt finance, if the token spend, instead of doing this, just does that, you've got a problem. If it were equity finance, it's a different thing. But because so much of it is debt financed, and that token spend, going from this to that could come from open weight, which we've already talked a lot about, but also from local compute. I think that's the part that people are not appreciating, is the ability to run the models on an air gap computer based on local compute.

Scarlet Fu (3:55)

You're not drawing in the cloud.

Tom Keene (3:56)

Can I do a family shout out?

Scarlet Fu (3:57)

Please.

Tom Keene (3:57)

My brother's dead on in this local weight. He says you're 100% right.

Peter Orszag (4:01)

On local compute.

Tom Keene (4:02)

On local computing.

Scarlet Fu (4:03)

So there's so much we don't know yet about how AI is going to work and how it's going to affect the economy. One thing we do know is what Tom referenced, which was your work on the research that led to automatic enrollment in 401ks. Your view on the Trump accounts, which have now made clear that those kids under 18 will now be automatically enrolled in these Trump accounts. Is it a good thing?