Someone on the internet said that to get people to read your academic papers, you just had to put Harry Potter and in front of the title.
Another thing that seems to work for attention is applying the AI lens:
How Will AI Affect the Way Snails Perceive Color?
The AI Perspective on Knitting Hats for Tangerines
Is It True that AI Generates Four out of Five Dentists’ Flossing Recommendations?
And while I am certainly not implying that Michael Barr, the Fed’s main architect of the now indefinitely suspended US Basel III endgame, would ever use such tricks to entice an audience, he did recently give a speech on Artificial Intelligence in the Economy and Financial Stability.*
The speech is all about different hypothetical scenarios, incremental progress, and transformative change, for the economy’s adoption of generative AI and how it may impact financial stability which is, of course, interesting in itself.
(Spoiler alert: AI could both stabilize and destabilize the financial system depending on how we govern and regulate its use.)
The true rabbit hole, however, is a CEPR/VoxEU column titled AI Financial Crises* that Barr references as he talks about AI and financial risk.
I know that many of you have looked at the CBOE Volatility Index (VIX) lately and thought: Wow, that is not as asymmetrically spikey as it used to be!”
No? Well, maybe that’s just me then.
But if you did do that, you are right. Specifically, while spikes might not be fewer or smaller, they tend to disappear much faster now than ten or twenty years ago. And what are spikes in volatility but financial risk!?!
The CEPR/VoxEU column lists some interesting reasons for that. And one of the ways that AI can exacerbate financial instability is through the oligopolistic market structure channel. (I know, it was a long walk, but now we are here.)
Because there are not that many tech companies that design and run AI engines—they have an oligopoly—financial institutions all rely on similar engines and models that are trained on pretty much the same data and therefore react to volatility in the same manner.
The downside is that the reaction comes faster and more forceful than before. A run to sell risky assets and buy safe-haven assets can happen in minutes if not seconds.
The upside is that the reaction is internalized just as fast, and therefore the system can recover in the blink of an AI, too.
Said in a less-cheeky way, AI may cause herd mentality (risk monoculture) in financial markets but that works for the freak out and the calm down alike, i.e., the spikes are (more) symmetrical.
*You can read Michael Barr’s speech here and the CEPR/VoxEU column here.
Regitze Ladekarl, FRM, is FRG’s Director of Company Intelligence. She has 25-plus years of experience where finance meets technology.
This article is part of the FRG Risk Report, published weekly on the FRG blog. To read other entries of the Risk Report, visit frgrisk.com/category/risk-report/.