The 3 major decisions the next Congress must make about AI policy 

Imagine: In a shocking turn of events following the November elections, the U.S. Congress has failed to enact crucial AI policy, leaving the rapidly evolving technology sector in regulatory limbo. Industry insiders warn of potential chaos as AI-driven systems proliferate unchecked across various sectors. Critics argue that legislative inaction has opened the door for tech giants to consolidate power, with AI-assisted lobbying efforts reportedly overwhelming human policymakers.

Meanwhile, concerns mount over the deployment of sophisticated AI in political campaigns, raising fears about the integrity of future elections. As the AI arms race accelerates, experts caution that, without proper oversight, the technology could fundamentally alter the balance of power in both the private and public sectors. 

Is this what the American people want? 

During the 118th Congress, more than 100 artificial-intelligence-related bills have been introduced. Not a single one has passed. Yet 59 percent of Americans believe elected officials are paying “too little” attention to issues dealing with technology and technology companies, according to a Gallup survey from 2020.  

What has them concerned? Take, for example, AI-powered discrimination in the workplace, deepfake child sexual abuse material and fraudulent robocalls, to name a few. While states have introduced legislation to tackle some of these specific harms, there has been no such action in Congress.   

With all eyes on the presidential race, it’s really Congress that we should be keeping an eye on when it comes to AI policy. The makeup of the next U.S. Congress will have massive implications for the development of AI. The race to establish the mightiest AI system is not just a race between Silicon Valley tech companies, but one between countries — especially the U.S. and China — competing for global supremacy.  

Having established a Select House Committee on the Chinese Communist Party and a bipartisan Senate working group on AI, U.S. policymakers are grappling with and increasingly interested in these topics. In 2020, Congress did manage to pass the National Artificial Intelligence Act, which established the American AI Initiative. Since then, interest has been high, but action has been scarce. But it’s no longer just about interest. It’s about necessity.  

The time for Congress to act is now. Why? Because no one else can.  

Up until now, AI policy has been largely led by the executive branch. This goes all the way back to 2016, when the Obama administration’s National Science and Technology Council released a report that emphasized AI’s vast potential to open new markets and improve health, education, energy and the environment.  

This was followed by the Trump administration’s focus on doubling AI research investments and establishing research institutes, driven by a belief that AI is critical to U.S. economic and national security.   

The Biden administration marked a shift to comprehensive attempts at AI policy. President Biden issued an expansive executive order stirring federal agencies to act and develop their own AI research, standards, guidelines and policies. His administration also elicited voluntary commitments from the biggest AI companies and stepped up international engagement efforts.  

However, the Supreme Court’s decision to overturn the Chevron deference — a doctrine that allowed courts to defer to federal agencies’ interpretations of ambiguous laws — signaled a major shift. We have yet to see the full legal repercussions of this play out, but for now, agencies will have less appetite to step into their regulatory roles in the face of uncertainty.    

In the absence of nationwide policies, state legislatures have become more active, with over 45 states introducing AI bills, 31 of which enacted legislation. States like California have taken the lead and stepped into the regulatory void, creating a patchwork of AI-related laws that will lead to legal battles over jurisdiction and authority.  

This fragmented approach risks stifling innovation while leaving major ethical and safety concerns unresolved. The only way to stitch this patchwork is through Congress.  

The 119th Congress will have to make three crucial decisions about AI:  

1. What values will guide our AI strategy? Our policy must address national security, market efficiency, power concentration, AI safety concerns, industrial policy and the public interest. Only elected officials can legitimately represent the voice of the American people in balancing these needs. 

2. Should long-term existential and catastrophic risks shape a regulatory focus on the technology itself, or should we prioritize addressing existing harms and biases through clear liability frameworks and regulations on the use of the technology? 

3. Do we need a new AI or Digital Technologies Agency, or should the regulation, supervision and enforcement of AI-specific statutes be delegated to existing agencies? 

The path forward for AI policy is not a dystopian nightmare but a challenge we can meet with thoughtful, decisive action. By engaging in robust debate and making informed choices, we can harness AI’s potential while safeguarding our values and democratic principles.  

The questions before us are complex, but not insurmountable. Whether through existing frameworks or new institutions, we have the power to shape AI’s impact on our society. The key lies in our commitment to balanced, forward-thinking policies that prioritize both innovation and the public good. With careful consideration and bold leadership, we can ensure that AI becomes a tool for progress, not a threat to our future. 

Slavina Ancheva is a Master in Public Policy candidate at the Harvard Kennedy School, Fulbright Bulgaria Scholar, and Belfer Young Leader. Paulo Carvão is a former IBM executive and senior fellow at Harvard’s Mossavar-Rahmani Center for Business and Government.