What are agencies doing with AI? A bipartisan Senate duo wants to know.
Sens. Ron Wyden, D-Ore., (L) and Mike Crapo, R-Ida. lead a Finance Committee hearing in March 2022. Drew Angerer/Getty Images
Sens. Ron Wyden, D-Ore., and Mike Crapo, R-Idaho, sent letters to three federal agencies probing into their use of artificial intelligence systems.
Lawmakers on the Senate Finance Committee are asking for clearer information on how some federal agencies use artificial intelligence technologies in their operations to ensure that such uses are safe and secure.
Sens. Ron Wyden, D-Ore., and Mike Crapo, R-Idaho issued letters to the Department of Homeland Security, U.S. Customs and Border Protection and the Department of Health and Human Services to better understand if, and how, these agencies are employing automation in their missions.
“As is frequently the case with new technology, AI provides us with exciting opportunities to better serve the American people, but we’re only beginning to see the consequences of leaving these systems unchecked,” Wyden said in a statement. “The federal government has a responsibility to ensure the systems it is using to make decisions that impact Americans’ daily lives are doing so accurately and without harmful bias.”
In the letter sent to DHS Secretary Alejandro Mayorkas and CBP Commissioner Troy Miller, Wyden and Crapo outline the various use cases AI systems can have in these agencies’ missions, like catching anomalies in customs documentation and leveraging advanced screening to prohibit illegal drugs from entering the U.S.
The lawmakers are seeking insights into how exactly CBP is leveraging AI in its business operations, with an emphasis on controlling for biases and conducting cost-benefit analyses for each AI tech acquisition.
“Without proper guardrails, CBP’s use of AI could result in trade facilitation and enforcement systems that are ultimately less effective,” the letter reads.
Similarly, the letter sent to HHS Secretary Xavier Becerra questions how any AI systems process and handle sensitive data from HHS. The lawmakers also want information about AI systems used in clinical care and how transparent these technologies are when handling cases in Medicare, Medicaid, child welfare and more.
“As the Senate Committee with jurisdiction over Medicare, Medicaid, the Children’s Health Insurance Program and federal child welfare programs that together cover over half of the U.S. population, we have a responsibility to ensure access to quality health care and services for beneficiaries, particularly given the vital role that federal health care policies play in supporting the adoption and reimbursement of safe and effective new private sector technologies,” the letter reads.
Earlier this year, Wyden also sent a letter to the IRS asking similar questions on AI oversight.