Biden staff are meeting regularly to develop AI strategy, White House says

The White House chief of staff’s office is meeting multiple times a week to develop ways for the federal government to ensure the safe use of artificial intelligence, the Biden administration said on Tuesday.

As quickly emerging AI tools and chatbots capture attention across the country, the meetings, which the White House said take place two to three times a week, are the latest piece of the administration’s push to understand and eventually regulate the wide-ranging impacts the technology has on the country. It is also seeking commitments from tech companies on combating potential challenges, according to a White House news release.

“We’ll see more technological change in the next 10 years than we’ve seen in the last 50 years, and maybe beyond that. And AI is already driving that change in every part of American life,” President Joe Biden said on Tuesday afternoon from San Francisco, where he met with academics and experts to discuss both positive and negative consequences of AI. “We need to manage the risks to our society, to our economy and our national security.”

Professor Fei-Fei Li, one of the experts who participated in the meeting, described the conversation as “extremely engaged,” with the president “not just listening but asking questions.”

“It was very, very balanced and thoughtful,” Li said about the tone of the conversation with Biden.

“There was recognition of how the technology, if not used responsibly, could have very negative implications and how it could have very constructive and positive implications,” Li told POLITICO, adding that she had urged Biden to take a “moonshot mentality” in terms of investing in the public sector to ensure responsible “stewardship of the technology into society.”

The Biden administration’s newest steps will build on the existing “Blueprint for an AI Bill of Rights,” a set of principles released in October that seek to guide the administration’s approach to ensuring privacy and equity alongside the use of the technology. These guidelines warn against algorithmic discrimination, urge data protection and underscore human alternatives to AI abilities. Biden has also signed an executive order that directs federal agencies to rid new technologies of bias.

The pressure for the federal government to regulate AI has ramped up particularly in the past year, most notably after OpenAI’s release of ChatGPT in November. The chatbot, which reached more than 100 million active monthly users two months after launch, can perform a wide array of tasks, from passing accreditation and standardized exams to writing emails and computer code. OpenAI’s founder, Sam Altman, called on Congress to regulate AI in a Senate hearing last month.

Alongside ChatGPT, other generative AI platforms with the capability to create photos, audio recordings and video clips have also seen increased use in the past few months. Experts have said the AI platforms that have seen a rapid rise could aid worker productivity, but at the same time kill jobs with their competencies.

The White House’s move toward setting guardrails on AI contrasts with a lack of action on Capitol Hill to regulate the fast-moving technology. Despite the topic receiving wide attention, members of Congress across the political spectrum, from Rep. Ted Lieu (D-Calif.) to Sen. Ted Cruz (R-Texas), have questioned the ability of lawmakers to enact appropriate legislation, citing a general lack of expertise on the topic and potential outcomes of laws.

Lieu and Rep. Ken Buck (R-Colo.) are introducing a bill to form a commission that will strategize around regulating AI, according to The Washington Post.

The U.S. is also playing catch-up internationally on the topic, as China and the EU have either passed legislation or set rules putting guardrails on the use of AI. Individual states have already pushed to regulate AI, including Colorado, which passed a law in July 2021 that requires insurance companies to test algorithms for bias.

In the political realm, the prospect of fake content has created concerns among campaign strategists who fear the dissemination of misinformation. One Twitch livestream of an AI-generated debate between Donald Trump and Biden in which users can interact has garnered the attention of publications like Yahoo in the past few days.

Mohar Chatterjee and Brendan Bordelon contributed to this report.