Webmaster's Home (ChinaZ.com) July 18 news: The US Securities and Exchange Commission (SEC) expressed concern about the impact of generative artificial intelligence on financial markets.
Gary Gensler, chairman of the US SEC, said in a speech that generative artificial intelligence may increase the risk of financial markets because it may cause financial institutions to use inaccurate or irrelevant information to make wrong decisions.
Gary Gensler stated that his biggest concern is that all "downstream players", i.e. retail investors, venture capital firms, advisors, etc., are likely to get the same financial information and advice from very few underlying generative AI models. This would lead to a "monoculture" of the economy, and if everyone made decisions based on the same mortgage data and the housing market crashed, the entire economy would be at risk. Regulation could be based on individual models as well as the AI ​​field as a whole, but that's just an investor-side issue, he said.
Another concern is how bad actors can misuse generative AI for large-scale deception. In his presentation, Gensler mentioned a text from an AI-generated bot Twitter account promoting rumors that he had resigned.
Gensler also shared his concern that AI will become a new tool to confuse naive people in the next large-scale financial fraud scheme, as campaigns can now be tailored to individuals based on personalized AI algorithms. Insurers could use the models to determine who is more likely to have access to medical treatment, Gensler said, which may have moved beyond the dark side of machine learning algorithms.
The SEC is not new to concerns about the impact of artificial intelligence, having established FinHub in 2018 to answer questions related to artificial intelligence, cryptocurrencies and other fintechs. Current risk management guidance needs to be updated to accommodate the new technology, Gensler said, but may require a rethink of how it is used across the financial industry. Other government agencies have also begun investigating AI companies, such as the FTC's investigation of OpenAI.
