SEC Intensifies Scrutiny on AI Disclosures, Warns Against Misleading Information
On September 4, 2024, the U.S. Securities and Exchange Commission (SEC) Chair, Gary Gensler, once again expressed his concerns regarding artificial intelligence (AI)-related disclosures, underscoring the need for public companies to provide honest and comprehensive communication concerning their AI use and technological capabilities. Gensler's comments come against a backdrop of recent cases pursued by the SEC and escalating shareholder class action litigation that allege misrepresentation regarding companies' utilisation of AI in their operations, business, and revenue practices.
Gensler's remarks were made during an "office hours" video session, where he cautioned public companies, investment advisers, and broker-dealers against what he termed "AI washing," a practice where entities make misleading claims about their AI capabilities. This echoes similar warnings he made in December 2023. Gensler emphasised the necessity for AI disclosures to reflect "material risk" specific to the company rather than relying on generic templates. Furthermore, public companies may be required to define their definition of "AI," detailing how and where it is being used within their operations, and whether it has been developed internally or sourced externally.
Investment advisers and broker-dealers were also warned that making false claims about their AI utilisation could result in violations of securities laws. Gensler's statements came shortly before the SEC's Division of Examinations' October 2024 report that outlines the agency's ongoing examination priorities for 2025, including a continued focus on AI-related disclosures.
The SEC has actively pursued cases this year against companies for misleading AI-related statements. In March 2024, the SEC settled charges with Delphia (USA) Inc. and Global Predictions Inc., alleging both companies made false claims regarding their AI capabilities. Delphia was accused of fabricating claims about its use of AI in analysing client data for investment purposes, while Global Predictions was charged with falsely representing its use of expert AI-driven forecasts and claiming to be the "first regulated AI financial advisor."
Earlier this year, in February, the SEC also settled with Rockwell Capital Management LLC and its founder for failing to fulfil their promises to initiate a hedge fund utilising AI for predicting price behaviours. Furthermore, the SEC, in a joint action with the Department of Justice, lodged a complaint against the founder and CEO of a recruitment start-up alleging that false claims were made about the AI product's capabilities in job candidate matching.
Moreover, allegations of "AI washing" have increasingly found their way into securities fraud class action litigation under Section 10(b) of the Securities Exchange Act and Rule 10b-5. These cases often centre on misleading statements claiming exaggerated AI technical capabilities. For instance, in a claim against Oddity Tech Ltd., allegations were made that the company lacked viable AI technology, rendering their public statements about AI developments and validations misleading. Similarly, in litigation involving Upstart Holdings, Inc., plaintiffs alleged that pronouncements on the robustness of the company's AI-powered lending platform were deceptive, particularly as the platform could not handle changing interest rates. In another case, a court ruled that the plaintiff plausibly argued that Zillow Group, Inc. misled investors by failing to reveal difficulties with its AI-driven home pricing tools.
The SEC is currently soliciting public feedback on proposed amendments to the Securities Exchange Act of 1934 and the Investment Advisers Act of 1940. These changes are designed to tackle conflicts of interest when broker-dealers or investment advisers interact with investors through predictive data analytics technologies.
Gensler's recent remarks and the SEC's active proceedings indicate a sustained focus on ensuring transparency and accuracy in AI-related disclosures, compelling companies to assess and validate their communicated AI capabilities to avoid legal repercussions.
Source: Noah Wire Services