On November 19, 2025, Senators John Curtis (R-UT) and Mark Kelly (D-AZ) introduced the Algorithm Accountability Act, which seeks to amend Section 230 of the Communications Decency Act. This proposed legislation would require social media platforms to exercise reasonable care in the design and operation of their recommendation algorithms to prevent foreseeable bodily harm or death. Currently, Section 230 protects online platforms from liability for user-generated content and allows for good faith content moderation.

Under the proposed act, victims of algorithm-related harm would have the right to sue social media companies if they can demonstrate that the platforms failed to meet their duty of care. The bill specifically targets for-profit social media platforms with over one million registered users. The sponsors assert that the legislation would not infringe upon First Amendment rights, addressing common concerns regarding reforms to Section 230.

This legislative effort follows discussions about the role of social media in exacerbating political tensions and violence, particularly in light of recent events. Curtis has linked the toxic online environment to the radicalization of individuals, citing the case of conservative activist Charlie Kirk, who was killed in September. He argues that algorithms designed to maximize engagement may contribute to such violence.

The Algorithm Accountability Act bears similarities to the Kids Online Safety Act (KOSA), which has faced challenges in Congress due to lobbying and concerns over free speech. Notably, the new bill would not restrict the chronological display of information or limit users' access to content based on their viewpoints. Previous legal actions against platforms like YouTube and Meta have underscored the complexities surrounding algorithmic responsibility, particularly in cases involving hate speech and radicalization. Critics, including the Electronic Frontier Foundation (EFF), warn that the bill could lead to over-censorship by platforms to avoid liability, potentially undermining protections intended to safeguard users from harm.

In a separate development, President Donald Trump is considering an executive order that would grant the federal government significant authority over the regulation of artificial intelligence (AI). This order would establish an "AI Litigation Task Force" under the supervision of the Attorney General, tasked with challenging state laws perceived to hinder the growth of the AI industry. The draft of the order highlights California's recent legislation on AI safety and Colorado's law against algorithmic discrimination as examples of state regulations that could be contested.

Trump has expressed a desire for a moratorium on state AI laws, framing it as necessary to prevent what he describes as "woke" ideology from influencing AI development. He argues that navigating regulations across 50 states is impractical and could lead to inconsistencies. This perspective aligns with his broader AI Action Plan, which directs federal agencies, including the Federal Communications Commission (FCC), to find ways to bypass state regulations viewed as burdensome to the industry.

The proposed executive order outlines a 90-day timeline for key federal agencies, including the Department of Justice and the Federal Trade Commission (FTC), to implement its directives. Within this period, the Secretary of Commerce would be required to identify states that violate the federal AI policy and assess their eligibility for the Broadband Equity Access and Deployment (BEAD) program, which supports rural broadband initiatives.

FCC Commissioner Brendan Carr has suggested that the FCC could interpret the Communications Act to override state laws that impede the deployment of modern infrastructure, including AI technologies. He has raised concerns about state laws requiring AI companies to disclose their safety testing models, suggesting that such requirements could conflict with federal objectives.

The potential for the FCC to exert authority over state AI regulations, along with other elements of Trump's order, may face legal challenges. However, the establishment of the litigation task force could create obstacles for states attempting to regulate AI effectively. Reports indicate that this executive order may serve as a contingency plan if Congress does not pass a moratorium on state AI laws, particularly in the context of the National Defense Authorization Act, which is essential for government funding. Previous attempts to include a moratorium in Trump's proposed legislation faced bipartisan opposition, and similar challenges may arise if the moratorium is reintroduced in the NDAA. The effectiveness of leveraging federal funding to influence state lawmakers remains uncertain, particularly in larger states like California, which may resist such pressures.