Schiff: Congress Must Learn from Social Media Failures to Regulate AI Before It's Too Late

Senator Adam Schiff sits at the intersection of AI policy in Congress, holding positions on committees overseeing intellectual property, antitrust, privacy, and technology. In a recent conversation, the California Democrat expressed deep concern that Congress risks repeating its fumbling response to social media—but this time with even graver consequences.

The Corruption Backdrop#

Schiff opened by highlighting what he sees as pervasive self-dealing within the current administration. From campaign ads funded by taxpayer money to various "moneymaking schemes," he pointed to a pattern of governance driven by personal benefit rather than public interest. This backdrop frames his skepticism of recent moves in AI policy, including Trump administration actions framed as national security measures that lack binding force.

The Hollow Voluntary Pledge#

Tech executives recently gathered at the White House to sign a non-binding commitment to responsibly develop AI. Yet away from presidential cameras, the same leaders have publicly stated they support actual regulation, not voluntary pledges. Schiff sees this contradiction as revealing: executives fear the administration's retribution if they openly defy it, but privately acknowledge that self-policing is insufficient.

He emphasized that real national security threats exist in advanced AI development. As models become more capable and self-improving, their decision-making becomes less transparent and controllable—a risk he called "of the highest order." He also highlighted concerns over AI's role at the Pentagon, including questions about whether it played a part in targeting decisions in military operations, and noted administration efforts to undermine Anthropic for refusing to build AI systems for mass surveillance or autonomous weapons.

Learning—or Failing to Learn—from Social Media#

Schiff expressed frustration that Congress took years to grapple with social media's harms and still has little to show for it. The parallel challenge now is whether lawmakers will move faster on AI. Unlike social media, where regulatory efforts collided with First Amendment concerns around speech, AI regulation centers on input legality, output safety, and market competition—issues with clearer regulatory pathways.

He rejected the notion that the First Amendment itself blocked social media reform. Instead, he argued, tech companies strategically weaponized First Amendment rhetoric to avoid accountability, while Congress granted them near-total immunity under Section 230. That immunity was supposed to encourage moderation of genuinely harmful content; instead, platforms used it as cover for inaction.