New Poll: Americans Want Trust, Accountability, and Independent Expertise as AI Guardrails

Fathom's latest national survey finds durable, bipartisan demand for AI governance built around verifiable standards, accountability for harms, and independent expertise

WASHINGTON, April 17, 2026 /PRNewswire/ -- A recent national report released by Fathom, AI Governance: What Americans Really Want, finds that Americans across party lines, age groups, and levels of AI familiarity consistently prioritize trust, accountability, and independent expertise as the foundation for how AI should be governed. The findings come as state lawmakers consider legislation to establish a framework of independent verification for AI products.

Unlike standard opinion polling, which tends to surface easy agreement on abstract principles, Fathom's survey was designed to test where the public's preferences hold once real-world tradeoffs are introduced. Respondents were asked to rate twenty principles for a good AI future, then to weigh those same priorities against explicit costs — accountability "even if this creates liability risks for companies," verification "even if it slows innovation" — and finally to make either/or choices between competing principles. The result is a durable portrait of public opinion that can guide lawmakers.

Key findings include:

• Trust and accountability are the organizing principles of public opinion on AI. Child safety leads at 90% "very important," followed by accountability for AI-caused harms, verifiable standards, and preserving human reasoning. When tradeoffs are made explicit, support barely moves: trust and verification hold at 86% total importance, accountability at 87%.
• Independent experts are the most trusted voice in AI governance. 71% of Americans trust independent experts to lead the AI transition, ahead of tech companies (61%), federal agencies (51%), and elected officials (37%). This hierarchy has held across all three waves of Fathom's polling, and the trend is strengthening.
• The public decisively rejects a "let the market sort it out" approach to workforce disruption. Every workforce intervention tested received majority support, from retraining programs (84%) to a sovereign wealth fund sharing AI-generated wealth with the public (71%).
• Consensus thins where governance collides with American competitiveness. While 83% support slowing frontier AI development until risks are better understood, support drops sharply when slowing down requires the U.S. to give up control over its own AI decisions to international coordination.

"Americans across the country want AI that they can trust. As agentic AI systems become increasingly integrated in our financial systems, hospitals, workplaces, and daily lives, the risks from opaque, unaccountable, and untrustworthy systems will only grow," said Andrew Freedman, Co-Founder and CEO of Fathom. "This polling shows that the demand for guardrails is durable, and it points clearly to what kind of governance the public wants: verifiable standards, clear accountability when things go wrong, and independent experts at the center of oversight rather than companies policing themselves."

The report identifies four takeaways for policymakers: build governance frameworks on the durable consensus around trust and accountability; frame trust and accountability as competitive advantages rather than constraints on innovation; act on workforce policy while public preferences remain flexible; and give independent experts and civil society a meaningful seat at the oversight table.

This report builds on Fathom's polling from the fall, which finds that Americans are using AI more than ever — but are deeply skeptical of both government and industry to oversee it. Only 35% trust elected officials to govern AI, while nearly 80% support guardrails on key priorities including child safety, privacy, and human oversight. A companion report found that 67% view the Independent Verification Organization (IVO) framework favorably as a concrete path forward beyond government regulation or self-regulation alone. The full reports are available at https://fathom.org/insights.

About Fathom: Fathom is an independent nonprofit whose mission is to build a governance architecture that helps society navigate the transition to a world with AI by fostering trust, safety, and innovation. Learn more at fathom.org.

SOURCE Fathom AI Inc.