Conservative provocateur Steve Bannon and progressive Senator Bernie Sanders are joining a chorus of voices, including lawmakers, religious leaders, and actor Joseph Gordon-Levitt, to express significant concern over the regulation of artificial intelligence. This unexpected alliance highlights a growing mainstream anxiety about AI development.

The concerns surrounding advanced AI models have been escalating for months as these systems have begun exhibiting unpredictable behaviors. The issue gained significant public attention last week when a researcher at the AI company Anthropic publicly resigned, warning that AI could pose an existential threat by the end of the decade. This sentiment, once confined to science fiction, has moved into the realm of serious discussion, with leading AI researchers and executives echoing calls for measures to curb the technology's rapid advancement.

A "Pro-Human Assembly" event in Washington is set to feature Sanders, Bannon, Republican Senator Marsha Blackburn, Congressional Progressive Caucus chair Greg Casar, and the Archbishop of San Francisco, among others, all advocating for stronger AI safeguards.

However, former President Donald Trump has expressed strong opposition to the idea of an AI slowdown, describing it as a "SICK conspiracy" that benefits China. His stance suggests a significant political hurdle for any potential AI regulation.

Potential Regulatory Approaches

Experts and advocates have outlined several potential avenues for government intervention to make AI safer. These approaches, while having historical precedent, face political challenges:

  • Bans or Pauses: Federal law could prohibit the development of AI systems that surpass human performance or pause their development until regulatory agencies are established to oversee them.
  • Empowered Auditing: A designated third-party authority, potentially a government agency, nonprofit organization, or industry group, could be tasked with reviewing AI systems before their public release. Previous efforts under the Trump administration were voluntary and secretive, with questions remaining about the technical expertise of reviewers.
  • Increased Liability: Congress could amend federal liability laws to hold AI companies accountable for damages caused by their systems. The threat of substantial lawsuits could incentivize companies to adopt more cautious development practices.

Despite the theoretical feasibility of these measures, the primary obstacle remains political consensus. There is a concern among AI researchers that the window for effective regulation may close while lawmakers struggle to align on a path forward, noting that Washington, D.C., operates at a much slower pace than Silicon Valley.