OPINION:
Artificial intelligence is here, and members of Congress are trying to use this development to make headlines. Politico reports that congressional leaders “are privately negotiating a deal to address increasing concerns about artificial intelligence, and they’re hoping to move a bill in the lame-duck period.”
Congress’ skepticism of AI is nothing new. Earlier this year, a congressional working group headed by Senate Majority Leader Charles E. Schumer released a report recommending any number of vague clampdowns on AI, such as establishing “appropriate guardrails” for the technology’s use in health care and considering legislation targeting deepfakes.
Irrespective of issues that may arise with artificial intelligence, the greater problem is legislative intelligence — or lack thereof. Congress doesn’t have the knowledge necessary to micromanage new technologies. Regarding AI, our legislators would be wise to recognize this and proceed with caution.
Congress’ knowledge problem on tech is on full display whenever a Silicon Valley CEO testifies before a committee. At a hearing in April 2018, Sen. Orrin Hatch, who had been in office for more than 40 years, asked Facebook President Mark Zuckerberg how his company made money. (“Senator, we run ads,” Mr. Zuckerberg smiled.) That same year, Rep. Steve Cohen told Google’s CEO, “I use your apparatus often.”
How did Congress “get dumb on tech,” as the Washington Monthly put it a few years back? The answer is that modern technology is complicated, requiring a staggering degree of fluid knowledge that Congress doesn’t and can’t have. Legislators are occupied with many issues; they don’t have time to specialize in something as demanding as AI.
The same might have been said of radio, a newish technology in 1927 when Congress passed the Radio Act. The legislation created a regulatory commission that would later become the Federal Communications Commission and tasked it with ensuring broadcasts were in the “public interest.” This censorious mandate eventually snowballed into the fairness doctrine, which was supposed to create a balance of viewpoints only to backfire and chill free speech for decades. The fairness doctrine was eventually repealed but was a direct result of a Congress that had overreached on a technology it didn’t understand but sought to control anyway.
Radio is far from the only technology where lawmakers has shown a lack of imagination. In the 1960s, California state legislator Nicholas Petris began a push to ban cars with internal-combustion engines, citing the pollution they emit. Congress got involved, and several similar bills were introduced at the national level, one of which would have phased out gas engines within three years.
Today, gasoline-powered cars are vastly cleaner than in Petris’ time. While some of this concerns government emissions standards, much of it can be credited to technological innovation and consumer demand.
Today’s congressional designs on AI are likewise doomed to prove shortsighted. As one speaker at an AI symposium recently said, lawsuits filed over artificial intelligence “will look completely juvenile and ridiculous” two years from now.
Pattern-recognizing algorithms used in AI have been employed for decades in everything from HOV lanes to dynamic pricing on hotel rental websites. But that has not stopped Congress and the Justice Department from going after RealPage and Yardi (property managers’ AI-based property management software), hotels’ pricing algorithms and others.
AI has caught on for a reason: It makes people’s lives easier, including radiologists and police officers. Close to three-quarters of American companies have adapted AI for at least one business function; the AI industry is expected to see annual growth of nearly 37% from 2023 to 2030. Companies such as Dell Inc. have launched divisions such as the Dell AI Factory to assist businesses in making informed decisions regarding how AI can transform business processes to give a competitive edge.
Artificial intelligence is key to the future of our economy and national security.
The United States is in what is called an AI arms race with China. This arms race has the power to reshape the global economy. If America wants to stay on top of its economic rivals China and Japan, it must lead innovation and development rather than hinder the tech sector with regulations.
Technology arms races between the U.S. and China are nothing new, and we have seen similar congressional struggles in the race to develop 5G technology. The 5G race was mostly led by the private sector with a clear understanding that Congress did not fully understand the issues (other than that the U.S. must win the race) and that adequate political intelligence did not exist on Capitol Hill to “tackle the complicated, multifaceted and resource-intensive issue of 5G in a meaningful way.”
Hypothetical horror stories about real-life Terminators are far more likely to impede prosperity than become reality. A new study finds that Americans have adapted to generative AI much faster than they did to the personal computer or the internet. This suggests that technology isn’t as disruptive or dangerous as some fear.
The sectors that readily adopt AI (health care, finance, insurance, transportation, law enforcement, energy, manufacturing, banking) are already federally regulated. The regulations already have frameworks and rules in place to protect the public and require systems to be certified against certain standards before customers can use them. These regulations require due diligence before the adoption of any new technologies and will go far in protecting consumers and the public from any new technological innovations — including AI.
Thankfully, some in Congress are catching on. As Sen. Dick Durbin, Illinois Democrat, put it, “When you look at the record of Congress in dealing with innovation, technology, and rapid change, we’re not designed for that.” Sen. Ted Cruz, Texas Republican, agrees, saying of AI, “To be honest, Congress doesn’t know what the hell it’s doing in this area.”
AI is developing quickly, and available information on AI is changing weekly, too fast for nontechnical political groups to understand. Rather than do too much, Congress should proceed cautiously. Let markets and the enlightened choices of stakeholders and individuals rather than government fiat determine the future of AI.
• Adam Scott Wandt is associate professor of public policy and vice chair for technology in the Department of Public Management at John Jay College of Criminal Justice. He is a co-chair of the New York City Bar Association committee on Technology, Cyber and Privacy Law.
Please read our comment policy before commenting.