WASHINGTON — President Biden issued a sweeping executive order Monday regulating the development of artificial intelligence — after Elon Musk, Sam Altman and other industry leaders have warned the technology’s unsupervised advancement could pose a risk to humanity.

Under the first-of-its-kind action, companies such as Google, Amazon and Microsoft-backed OpenAI will be required to share safety test results with the government whenever they train an AI model that poses a potential “serious risk to national security, national economic security, or national public health and safety.”

The 80-year-old president said before signing the order at the White House that there were “challenges and opportunities” and that he wants to ensure American dominance of the sector while limiting the risks.

Biden specifically said he wants to prevent AI from making social media “more addictive” or from abetting fraud — noting that he’s seen a “deep fake” video of himself, using a term for the increasingly prevalent and convincing doctored videos.

“President Biden is rolling out the strongest set of actions any government in the world has ever taken on AI safety, security, and trust,” White House deputy chief of staff Bruce Reed said in a statement Monday morning.

“It’s the next step in an aggressive strategy to do everything on all fronts to harness the benefits of AI and mitigate the risks,” he added.

Senate Majority Leader Chuck Schumer (D-NY) attended Biden’s event and said afterward that the president’s actions could help prevent the elimination of mankind — though he said Congress must adopt even more rules.

Asked about 42% of CEOs at a recent forum expressing concern that AI could “destroy humanity in 10 years,” Schumer said on the White House driveway: “That is one of the reasons the president did invoke the [Defense Production Act], to make sure there is testing to avoid that.”

“Obviously doomsday scenarios worry people and we have to guard against them,” Schumer said.

However, the senator added, “Lots of the things that were done here can only be done with federal contracts … For most of this, legislation is required to cement it, to expand it, to make sure everybody obeys it — not just people who have contracts with the government.”

The order largely seeks to expand upon voluntary cooperation from leading technology companies, task key government agencies with making various reforms, and lay out additional guidelines for federal use of AI. 

Biden is giving most agencies between three months and one year to fulfill their part of the executive order, administration officials told various outlets. 

Underpinning the executive order is a push for the government to develop stronger standards for the use of AI.

One area that will receive particular oversight is the use of AI to develop biological materials after some AI experts expressed fears that the technology could be used to help engineer bioweapons.

The order requires federal agencies that bankroll projects to develop standards for use of AI in order to be eligible for funding.

Biden is also tasking the National Institute of Standards and Technology to craft metrics for so-called “red-testing” to ensure safety in the programs.

To help consumers distinguish between content built mostly by humans from that constructed by AI, the Commerce Department will set up guidelines for watermarking AI-generated content.

Additionally, the president is tapping the National Security Council to iron out ethical AI principles for use of the technology in intelligence-gathering operations.

To safeguard privacy, Biden is calling for the federal government to help finance research into technologies like cryptographic tools that can protect against eavesdropping.

The administration is also planning to study how companies extract commercially available information and encourage the use of AI training modules that are designed to protect privacy.

The Department of Justice and federal civil rights offices will also be required to compile best practices on how to prevent the technology from being used to amplify discrimination.

For workers, the administration will write up a report on ramifications of AI on the labor market and try to prevent employers from using AI to undercompensate their employees.

Looming over the new EO are concerns that bolstered AI regulations could stifle US development of the critical technology and cede ground to geopolitical rivals such as China.

To address this, Biden is making the State and Commerce departments push for international frameworks on the new technology.

The administration will also work to streamline visas to entice highly skilled workers and developers into the US.

Additionally, the administration will build up the National AI Research Resource, which will help give AI developers access to an array of resources to further their research projects in a bid to buoy innovation.

Critics of the regulatory push argue that Big Tech leaders are taking advantage of public fears to push for rules that will effectively fence out competition within the burgeoning industry.

Carl Szabo, vice president and general counsel of NetChoice, called the executive order “the wrong approach to govern AI” and “dangerous for our global standing as the leading technological innovators.”

NetChoice is a coalition of trade associations, ecommerce businesses such as Amazon and Google, and online consumers, “all of whom share the goal of promoting convenience, choice, and commerce” on the internet, according to its website.

“AI has vast potential to improve people’s lives, and we must not allow fears to hold the United States back,” the site also states.

“Broad regulatory measures in Biden’s AI red tape wishlist will result in stifling new companies and competitors from entering the marketplace and significantly expanding the power of the federal government over American innovation,” Szabo said. “Thus, this order puts any investment in AI at risk of being shut down at the whims of government bureaucrats.”

“There are many regulations that already govern AI,” he added. “Instead of examining how these existing rules can be applied to address modern challenges, Biden has chosen to further increase the complexity and burden of the federal code.”

Biden’s executive order comes as Congress slowly stitches together different frameworks to regulate nascent technology.

Schumer has huddled with industry leaders and convened forums to chart a path on regulating AI, though specifics have been limited. 

Other lawmakers have pitched competing concepts. 

Sens. Richard Blumenthal (D-Conn.) and Josh Hawley (R-Mo.), for instance, have unveiled a blueprint that would create a federal licensing and liability system as well as mandate transparency. 

Meanwhile, Senate Minority Whip John Thune (R-SD) and Sen. Amy Klobuchar (D-Minn.) have been eyeing a light-touch approach that would likely require self-certification of safety in particularly risky AI programs, Politico reported. 

None of the proposed regulatory frameworks appear close to taking shape in legislation anytime soon.

The announcement of Biden’s executive order is the first in a series of AI-related events and initiatives planned for this week. 

Vice President Kamala Harris is due to jet off to the United Kingdom and take part in a Global Summit on AI Safety hosted by British Prime Minister Rishi Sunak starting Wednesday. 

That event marks the first meeting of its kind between Western allies and key industry leaders to grapple with the challenges and risks posed by AI.

WIth Post wires