Skip to content
AI.info

The Pulse

Senators propose disclosures for AI financing and investment risks

A bipartisan Senate bill would require public companies to disclose more about AI financing, governance and material risks. It would also direct the SEC to study AI accounting and report annually on automated investment strategies.

Senators propose disclosures for AI financing and investment risks

A bipartisan Senate bill would require public companies to disclose more about the financing arrangements and material risks behind their artificial intelligence investments. The AUDIT AI Act would also require disclosures about corporate AI governance and direct the Securities and Exchange Commission to examine AI accounting, according to a release from Sen. Mark R. Warner’s office.

Sens. Mark R. Warner (D-VA), John Kennedy (R-LA), Angela Alsobrooks (D-MD) and Mike Rounds (R-SD) introduced the legislation, Warner’s office announced on Nov. 19. The measure is a proposal, not an enacted requirement.

Financing under scrutiny

Non-financial public companies would have to disclose AI-related joint ventures, special purpose vehicles, variable interest entities and other major debt arrangements in SEC filings, at the commission’s discretion. The proposal addresses financing structures that can make it difficult to assess companies’ exposure to AI investments, according to Warner’s office.

Warner told POLITICO he was concerned about circular investments, in which AI companies financially back firms that are also their customers. He cited Nvidia’s support for businesses that buy its chips and said more information was needed about financing through entities that might not appear directly on a company’s books.

Governance and risk reports

The proposal would also require public companies to say whether they have an AI governance policy and a dedicated board committee focused on AI, and explain why if they do not. Companies that have an AI policy would have to make it public.

Companies with significant AI exposure would have to provide standardized disclosures of material risks. These would include dependence on outside models, concentration of computing resources, costs and data liabilities.

The bill would direct the SEC to study how companies account for AI research and development, revenues and capital expenditures, then recommend future policy or standard-setting measures. A separate annual SEC report would examine how automated investment strategies affect price discovery, market efficiency and potential herding behavior.

The investment backdrop

A Brookings Institution research summary projects that investment in data center buildings, power systems, networking infrastructure, specialized chips and other equipment will total $10.3 trillion from 2025 to 2032, averaging 3.63% of U.S. gross domestic product per year. The projection comes from a conference draft by Stijn Van Nieuwerburgh, a professor at Columbia Business School.

The paper warns that financing risks are shifting from major corporations’ on-balance-sheet financing toward less transparent arrangements, including joint ventures, private credit, special-purpose vehicles, leases and loan guarantees. Van Nieuwerburgh cautioned that it would be premature to conclude that AI infrastructure already poses systemic risk comparable to earlier credit booms, while noting that off-balance-sheet structures may make correlated exposures hard to observe before a downturn.

The legislation was endorsed in Warner’s release by former SEC Commissioner Robert J. Jackson Jr. and former SEC chief accountant Paul Munter, as well as Van Nieuwerburgh. Warner told POLITICO that his aides had been in communication with SEC staff about the bill.

Sources

  1. Original source · Government or agency page Office of Senator Mark R. Warner warner.senate.gov

Other pages read

  1. POLITICO, republished by the Office of Senator Mark R. Warner warner.senate.gov
  2. Brookings Institution brookings.edu

Related stories

Explore

More articles