The Biden administration has secured voluntary commitments geared at managing the risks posed by artificial intelligence from seven top companies, the White House announced.
Amazon, Anthropic, Google, Inflection, Facebook parent company Meta, Microsoft and OpenAI have made the voluntary commitments, which include a pledge to develop mechanisms of watermarking to help users determine when audio and visual content is AI-generated, according to a White House official.
“Companies have a duty to do right by individuals and society and earn the American peoples’ trust by making it easy for users to tell whether audio and visual content is in its original form or has been altered or generated by AI, preventing bias and discrimination, shielding our children from harm and helping society overcome our greatest challenges, from cancer to climate change,” the official said.
The companies also committed to internal and external testing of their AI systems’ capability and safety before releasing them to the public, and to sharing results of those tests as well as information on managing AI risks across the industry.
Additionally, they’ve pledged to invest in safeguarding their models against a myriad of threats, to facilitate third-party discovery and reporting of their systems’ vulnerabilities and to publicly report their the systems’ capabilities and limitations, including security and societal risks.
Some of the companies have already been pursuing a few of the goals on their own, but the official described the new commitments as “pushing the envelope” on safety and security in the AI landscape.
President Biden convened the seven companies at the White House on Friday to announce the commitments. The move comes as part of the Biden administration’s effort to “seize the tremendous promise” of AI while also protecting Americans’ rights and safety against risks posed by the uncharted technology, according to a White House release.
“Americans are seeing how advanced artificial intelligence and the pace of innovation have the power to disrupt jobs and industries. These commitments are a promising step, but we have a lot more do together,” Biden said in remarks on Friday.
The president said that “in the weeks ahead” he’ll take executive action “to help America lead the way toward responsible innovation.” The White House says it’s developing an executive order that is related to responsible AI innovation, which the official called a “high priority” for Biden and his team.
Biden also called on Congress to pass bipartisan legislation that would put limits on data personal collection, ban targeted advertisements to kids and require companies to put health and safety first.
This story was updated at 2:16 p.m.