- SEC chief Gary Gensler shared his view on how artificial intelligence could impact markets.
- Tools like ChatGPT risk creating "monocultures" where every financial player uses the same models.
In an interview with The Messenger this week, Gary Gensler, the chairman of the Securities and Exchange Commission, warned of the risks artificial intelligence poses to financial markets.
While versions of the technology have already been in use for years in the industry, and various safeguards have been deployed for fraud and compliance, there are still ways AI can pose risks to investors' decision making.
"On the micro level, I think you're absolutely right that it can lead to bias, using AI to underwrite loans, insurance and the like," Gensler said at The Messenger's AI Summit: Balancing Innovation and Regulation in Washington, DC.
He said robo-advisers or broker-dealers, for example, have to be carefully programmed to optimize for the client's best interest, rather than for the advisor's interests.
But there are also systemic risks, Gensler said, stating that there could be a risk for financial companies and investors to all use the same few AI tools. This could have a "herding" effect in the market, something seen when groups or institutions all do the same thing or rely on the same tools, rather than acting independently.
"There are natural economics that will lead to monocultures, that there'll be base data sets or base models, and large parts of the financial sector will be relying on it...trading on it, underwriting on it," Gensler said. "Look, we've had a monoculture in FICO scores for decades so this is not an unusual thing or novel thing I'm saying."
But for people to rely on an AI model that has biases and shortcomings, he warned "you could have financial instability if you have large parts of the market relying on one data set on...they could kind of, the herding effect drive us off an inadvertent cliff."
Earlier this year, Gensler had similarly cautioned at the New York Times' DealBook conference that a companies will come to rely on a small handful of dominant AI models, which will eventually raise the odds of a crisis.
"This technology will be the center of future crisis, future financial crises," Gensler said in August. "It has to do with this powerful set of economics around scale and networks."