Senate Intelligence Chair Probes AI Companies
Summary
Senate Intelligence Committee Chair Mark Warner (D-VA) in mid-August sent a series of letters to CEOs of 13 AI companies—including the seven that committed to AI safety principles with the White House—urging them to take additional steps to ensure their products are deployed safely and used responsibly. The letters particularly voice concerns about the scope of the voluntary commitments and applicability only to more robust models instead of applying to models of all sizes, stating that “less capable models not covered by the commitments are susceptible to misuse, security compromise, and proliferation risks.” Sen. Warner also called on some companies that did not participate in the Administration’s announcement—including Midjourney, Scale AI, Apple, Stability AI, Databricks, and Mistral AI —to provide information on their efforts to prioritize safety, security and transparency. Chair Warner also recently sent a letter calling on Google to increase transparency, protect patient privacy and ensure ethical guardrails with regard to Med-PaLM 2—the company’s medical large language model.
