White House Announces Sweeping New Efforts to Protect Americans From AI’s Dangers

AI, of course, has also demonstrated vast potential for positive contributions, including to fight AI-assisted fraud — such as efforts by the Federal Communications Commission to use it to battle robocalls. The Department of Homeland Security (DHS) has implemented facial recognition and other AI tools for airport security, as well as to investigate border crimes. A few days ago when United Nations Secretary-General António Guterres announced the formation of an advisory body on the risks and benefits of AI, he cited “the transformative potential of AI for good,” along with serious concerns about “malicious use” of the tech.  

More from the White House order  

The order also requires:

• Tech developers to share safety test results on AI tech with the U.S. government, particularly when it “poses a serious risk to national security.”
  
• Congress to pass data privacy legislation and fund the development of cutting-edge privacy-preserving tech to protect all Americans, “especially kids.”
• Legislators and tech experts to develop best practices for AI to be used wisely and fairly in the justice system, and to prevent it from being used to discriminate in housing and other areas.
• Legislators to find ways to assist workers facing job loss due to AI, with federal support.
• DHS to establish an AI Safety and Security Board, which will bring together industry experts and government leaders “to help guide the responsible development and safe deployment of AI,” according to a statement from Secretary Alejandro N. Mayorkas, who will chair the board. 

Turning talk into action on AI safety

The alarm bells about AI began ringing last spring, when more than 1,000 tech leaders, including Elon Musk, signed an open letter warning that AI tools present “profound risks to society and humanity,” and urging a pause in companies’ development of the most powerful, advanced tech. “Should we let machines flood our information channels with propaganda and untruth?” the letter suggested we ask ourselves. “Should we risk loss of control of our civilization?”

A few months later, the White House elicited voluntary commitments from some major tech companies like Amazon, Google, Meta, Microsoft and OpenAI to double down on security efforts and publicly report any safety issues, as well as work to prevent AI-generated fraud and deception. 

This week’s announcement takes a firmer step toward regulating the technology — displeasing antiregulatory organizations like NetChoice, which issued a statement calling the order “Biden’s AI red tape wishlist” that “will result in stifling new companies and competitors from entering the marketplace.”

But a spokesperson for the Federal Trade Commission, which is focused on preventing a handful of leading AI companies from controlling the marketplace, praised the executive order in a statement to AARP as “a major step forward,” adding, “We’re encouraged to see it recognize the importance of a whole-of-government approach to promoting competition in AI."  

In general, tech insiders and advocates sound cautiously optimistic about the White House announcement. Karen Gullo, an analyst from the Electronic Frontier Foundation, which advocates for freedom and justice in the digital world, says she’s glad to see proposals for discrimination protections and “strengthening privacy-preserving technologies and cryptographic tools.” But, she adds, the executive order is “full of ‘guidance’ and ‘best practices,’ so only time will tell how it’s implemented.”