Lawmakers advanced a host of artificial intelligence bills Wednesday, marking Congress’ most significant step yet to address the technology’s risks and benefits, but emerging partisan disagreements could derail progress.

The Senate Commerce, Science, and Transportation Committee approved a number of bills that seek to boost AI research and development and establish voluntary guidelines on the technology.

Many of the panel’s Democratic and Republican members worked together over the past several months to introduce the legislation. Chair Maria Cantwell (D-Wash.) led at least three measures, which several Republicans co-sponsored and voted in favor of.

But ranking member Ted Cruz (R-Texas) blasted the effort as “overly broad” regulation that would hinder AI advancement and competition and protect the interests of major tech companies.

Both Democrats and Republicans on Capitol Hill have widely advocated for policies to strengthen AI development and maintain the US’ global lead while mitigating risks like displacing workers and perpetuating bias and discrimination. Senate Majority Leader Chuck Schumer (D-N.Y.) has called on committees to introduce and advance AI bills, and many lawmakers are eyeing to include such legislation in a year-end package.

Yet with Republicans like Cruz pushing for a more hands-off approach to AI regulation, the differences could threaten hopes for Congress to pass at least some AI bills by the end of this year.

AI ‘Roadmap’ Calls For Action on Key Bipartisan Bills, Policies

The tech industry, advocacy groups, and researchers have repeatedly called on lawmakers to respond to AI’s threats and opportunities and endorsed many of the bills approved in Wednesday’s markup.

Safety Institute

One of Cantwell’s bills (S. 4178), co-sponsored by Sen. Todd Young (R-Ind.), would establish the US AI Safety Institute—a coalition of government officials, tech companies, civil society organizations, and academics—to create voluntary standards on AI to promote safe use and development.

The bill would formally authorize the institute that the National Institute of Standards and Technology, a lab within the Commerce Department, has already been leading under President Joe Biden’s executive order on AI signed last October. Ahead of Wednesday’s markup, tech companies, including OpenAI Inc., Meta Platforms Inc., and Microsoft Corp., as well as AI policy groups and universities, were among nearly 50 stakeholders that urged the committee to prioritize enshrining the institute into law.

“It does not include any new requirement on companies,” Cantwell said of the bill. But, she added, it would support public-private partnerships and promote US competitiveness.

Cruz, a vocal critic of Biden’s executive order, rejected the proposal. He offered a failed amendment that sought to halt federal agencies from issuing AI guidance or rules without prior congressional approval.

“Private individuals develop this amazing product—all without the government’s help,” Cruz said. “Today, we hear without a trace of irony, that American AI can only be successful now with the government’s help.”

Gaining Traction

It’s a position that’s gained traction within some conservative circles critical of Biden’s AI strategy. The Republican National Committee’s 2024 platform, backed by presidential nominee Donald Trump, proposed to rescind Biden’s executive order if the party wins the election. The order directs federal agencies to establish guidelines and standards on AI to promote safety and security.

Republican Attack on Biden AI Policy Fuels Urgency on Guardrails

Cruz instead emphasized that Congress should advance “incremental” and “targeted” AI legislation, pointing to his bipartisan bill that seeks to combat the rise of AI-generated deepfake pornography as an example. The committee on Wednesday advanced the measure (S. 4569), known as the TAKE IT DOWN Act, which would outlaw the distribution of non-consensual deepfake porn and require websites to remove such content from their platforms within 48 hours of request.

The rest of the bipartisan AI bills lawmakers advanced were the:

• CREATE AI Act (S. 2714);
• TEST AI Act (S. 3162);
• Artificial Intelligence Research, Innovation, and Accountability Act (S. 3312);
• Promoting United States Leadership in Standards Act (S. 3849);
• National Science Foundation AI Education Act (S. 4394);
• Small Business Artificial Intelligence Training and Toolkit Act (S. 4487);
• Artificial Intelligence Public Awareness and Education Campaign Act (S. 4596); and
• Validation and Evaluation for Trustworthy (VET) Artificial Intelligence Act (S. 4769).