SEC Chair Says AI Could Cause the Next Big Financial Crisis

Tech hype comes in waves, and while the focus constantly changes (first crypto, then metaverse, and now AI), some things—like tech’s critics—remain the same. SEC Chair Gary Gensler has been the crypto community’s public enemy no. 1 for a few years now, but now he’s offering his somewhat caustic assessment on Silicon Valley’s latest heartthrob: artificial intelligence.

During a talk at the National Press Club conference on Monday, Gensler told the crowd that “AI may heighten financial fragility.” He offered some rather staid comments we’ve heard plenty of times before—that AI is the most transformative technology since the printing press—and so on. But he also said his agency was working to suggest new regulations that could stymie the potential for AI to abuse investors.

“AI may play a central role in the after-action reports of a future financial crisis,” Gensler said.

The biggest issue, he said, was that all the “downstream actors,” AKA all the retail investors, venture capitalist firms, advisors, and so on, could all be getting the same financial information and advice from very few base-layer generative AI models. This, he said, would create a “monoculture” of economics, which puts the entire economy at risk if, let’s say, everybody bets big on the housing market, which then tanks because everybody was basing their decisions on the same mortgage data.

Regulations could be based on individual models as well as AI as a whole, he said, but that’s just the investor end of the issue. The next is how bad actors can abuse generative AI for mass deception. The SEC head mentioned in his speech how a piece of AI-generated text from a bot Twitter account promoted the rumor that he had resigned from office. Gensler also shared his concerns about the way AI is just the next big tool for getting naive folks wrapped up in the next big financial fraud scheme, as now campaigns can be personalized to the individual based on personalized AI algorithms.

“With AI, fraudsters have a new tool to exploit… we all used to get spam, but it was all the same spam,’ he said. “Now communications can be individualized.”

Gensler should know about financial fraudsters. His agency has gone after some of the biggest lingering crypto companies like Binance and Coinbase. The SEC has also filed civil complaints against the alleged perpetrators of crypto-based fraud such as Celsius co-founder Alex Mashinsky and fuzzy-haired FTX ex-CEO Sam Bankman-Fried.

Rather than a large language model-based AI chatbot like ChatGPT, Gensler said the biggest use case for AI in making money is pattern recognition and its ability to make predictions about individuals, AKA the increasingly sophisticated models used to directly advertise and sell products to customers. If targeted ads weren’t already the dark side of machine learning algorithms, Gensler also mentioned how these models could be used by insurance companies to determine who might get access to medical treatment based on who is more likely to live.

He also shared his concerns that the algorithms themselves could simply reflect racial biases resulting from malformed training data. He said that AI models used to offer financial advice “must be in the best interest of clients and retail investors not… place the AI model’s interest ahead of the investors.”

Though crypto bros have considered the SEC as a main antagonist for years now, the agency chair took time before he fully became the Thanos-like figure to fintech tans. Though Gensler was originally “intrigued” by the nature of blockchain infrastructure as a “powerful force for good,” he eventually came to regard crypto as irrelevant, saying “We don’t need more digital currency.”

Perhaps Gensler’s opinion about AI will sour even more over time as the hype cycle wanes and the well-known limits of current AI become even more plain. But as with crypto, we’ll likely need to wait until the next big exploit drains investors of millions of hard-earned dollars before regulators truly decide to curtail this technology.

Want to know more about AI, chatbots, and the future of machine learning? Check out our full coverage of artificial intelligence, or browse our guides to The Best Free AI Art Generators, The Best ChatGPT Alternatives, and Everything We Know About OpenAI’s ChatGPT.

Tech hype comes in waves, and while the focus constantly changes (first crypto, then metaverse, and now AI), some things—like tech’s critics—remain the same. SEC Chair Gary Gensler has been the crypto community’s public enemy no. 1 for a few years now, but now he’s offering his somewhat caustic assessment on Silicon Valley’s latest heartthrob: artificial intelligence.

During a talk at the National Press Club conference on Monday, Gensler told the crowd that “AI may heighten financial fragility.” He offered some rather staid comments we’ve heard plenty of times before—that AI is the most transformative technology since the printing press—and so on. But he also said his agency was working to suggest new regulations that could stymie the potential for AI to abuse investors.

“AI may play a central role in the after-action reports of a future financial crisis,” Gensler said.

The biggest issue, he said, was that all the “downstream actors,” AKA all the retail investors, venture capitalist firms, advisors, and so on, could all be getting the same financial information and advice from very few base-layer generative AI models. This, he said, would create a “monoculture” of economics, which puts the entire economy at risk if, let’s say, everybody bets big on the housing market, which then tanks because everybody was basing their decisions on the same mortgage data.

Regulations could be based on individual models as well as AI as a whole, he said, but that’s just the investor end of the issue. The next is how bad actors can abuse generative AI for mass deception. The SEC head mentioned in his speech how a piece of AI-generated text from a bot Twitter account promoted the rumor that he had resigned from office. Gensler also shared his concerns about the way AI is just the next big tool for getting naive folks wrapped up in the next big financial fraud scheme, as now campaigns can be personalized to the individual based on personalized AI algorithms.

“With AI, fraudsters have a new tool to exploit… we all used to get spam, but it was all the same spam,’ he said. “Now communications can be individualized.”

Gensler should know about financial fraudsters. His agency has gone after some of the biggest lingering crypto companies like Binance and Coinbase. The SEC has also filed civil complaints against the alleged perpetrators of crypto-based fraud such as Celsius co-founder Alex Mashinsky and fuzzy-haired FTX ex-CEO Sam Bankman-Fried.

Rather than a large language model-based AI chatbot like ChatGPT, Gensler said the biggest use case for AI in making money is pattern recognition and its ability to make predictions about individuals, AKA the increasingly sophisticated models used to directly advertise and sell products to customers. If targeted ads weren’t already the dark side of machine learning algorithms, Gensler also mentioned how these models could be used by insurance companies to determine who might get access to medical treatment based on who is more likely to live.

He also shared his concerns that the algorithms themselves could simply reflect racial biases resulting from malformed training data. He said that AI models used to offer financial advice “must be in the best interest of clients and retail investors not… place the AI model’s interest ahead of the investors.”

Though crypto bros have considered the SEC as a main antagonist for years now, the agency chair took time before he fully became the Thanos-like figure to fintech tans. Though Gensler was originally “intrigued” by the nature of blockchain infrastructure as a “powerful force for good,” he eventually came to regard crypto as irrelevant, saying “We don’t need more digital currency.”

Perhaps Gensler’s opinion about AI will sour even more over time as the hype cycle wanes and the well-known limits of current AI become even more plain. But as with crypto, we’ll likely need to wait until the next big exploit drains investors of millions of hard-earned dollars before regulators truly decide to curtail this technology.

Want to know more about AI, chatbots, and the future of machine learning? Check out our full coverage of artificial intelligence, or browse our guides to The Best Free AI Art Generators, The Best ChatGPT Alternatives, and Everything We Know About OpenAI’s ChatGPT.