> ## Content Index
> Fetch the complete content index at: https://aiadoption.org/llms.txt
> Use this file to discover other available public pages before exploring further.

# 41% of Americans say AI regulation won't go far enough — only 27% say it will go too far
- URL: https://aiadoption.org/ai-analysis/41-of-americans-say-ai-regulation-wont-go-far-enough-only-27-say-it-will-go-too-far/
- Published: 2026-09-20T13:16:53.000Z
- Updated: 2026-09-20T14:13:09.000Z
- Description: 41% of Americans say federal AI regulation won't go far enough; only 27% say it goes too far, with concern strongest among adults 65+ and college graduates. In every US state, more people want more regulation than less.
- Author: Jassie
- Tags: AI Analysis, AI Compliance, AI Governance, AI Policy

If your reading of US public opinion on AI regulation assumes that the deregulatory pivot of 2025 reflects the dominant public preference, the 2025 Civic Health and Institutions Project (CHIP50) state-level survey shows the public position is the opposite. Across all 50 US states, 41% of respondents say federal AI regulation will not go far enough, versus 27% who say it will go too far. Roughly one-third say they are not sure. The federal executive orders pursuing AI deregulation and state preemption are operating against a public-opinion picture that consistently favours more, not less, AI regulation. The disconnect is structural and bears on what kinds of state-level legislation will retain public support and what kinds of federal positioning will face headwinds.

The state-level distribution shows the consistency of the pattern. In every US state, more respondents say AI regulation does not go far enough than say it goes too far. The states with the highest "not far enough" share are Missouri and Washington at 48%. The states with the highest "too far" share are New York and Tennessee at 31%. The uncertainty share (saying "not sure") is substantial across most states, typically 30-37%, making uncertainty the second-largest category in most states behind "not far enough."

*\[CHART fig\_936\_2026 — Attitude toward US AI federal regulation by demographic group, 2025\]*

The demographic breakdown is informative:

- 51% of adults 65+ say regulation won't go far enough, the highest of any demographic group.
- 46% of college graduates say not far enough, vs 34% of those with high school or less.
- 45% of Democrats say not far enough vs 40% of Republicans, a modest political difference.
- Women 41% not far enough vs 40% men.
- Rural 44% not far enough; suburban 36%; urban 38%.
- Income groups range from 39% (under $30K) to 46% ($70-99K).

The pattern: older, more educated, suburban-to-rural, middle-income Americans are the strongest constituency for more AI regulation. The "not far enough" position is bipartisan with Democrats slightly more represented. The "too far" position is similarly stable across demographics, between 25-35% across most groups.

The political-affiliation finding is the contested observation. Conventional framing treats AI regulation as a Democratic priority and deregulation as a Republican priority. The CHIP50 data shows the framing is incomplete. Republicans are nearly as likely as Democrats to say AI regulation won't go far enough (40% vs 45%). The 5-point gap is smaller than the gap between college graduates and high-school-or-less respondents (12 points), the gap between 65+ and 18-29 (around 14 points based on the demographic distribution), or the gap between college-educated and low-income groups.

**The implication: the federal deregulatory pivot is not operating from a strong public mandate.** The mandate framing requires that deregulation be the majority public preference; the data shows it is not. Federal AI policy in 2025-2026 is operating on a political coalition framework rather than a public-opinion framework. State-level legislation that responds to the public preference for more regulation has structural alignment with the actual majority sentiment.

## Three prescriptive moves follow for organisations operating against this regulatory environment.

**The first prescriptive move: plan for sustained state-level regulatory activity that responds to public preference.** The federal pivot toward deregulation does not eliminate the public demand for more AI regulation; it shifts the venue. State legislators responding to constituents will continue producing AI bills. The 150-bill 2025 state-level legislative pace is consistent with the 41% public preference for more regulation. The trajectory through 2026-2028 likely continues this state-level activity. Strategic compliance planning should engage with state legislative pace rather than federal deregulatory signalling.

**The second prescriptive move: position organisational AI practice ahead of likely future regulation rather than at the floor of current requirements**. The public preference is for more regulation; organisations that operate visibly above the current regulatory floor align with the public preference and bear less reputational risk if regulation tightens. Organisations that operate at the minimum required level bear the risk that regulatory shifts will catch them under-prepared and align them with the "regulation went too far" minority position.

**The third prescriptive move: engage with the older-adult, college-educated, middle-income constituency as the most influential regulatory voice.** The CHIP50 demographic data shows this segment has the strongest "not far enough" position. They are also the segment most active in legislative advocacy, civic engagement, and consumer voice. Communication and product strategy that addresses their concerns about AI explicitly (labour market effects, consumer protection, transparency, accountability) engages with the most influential regulatory constituency. Communication that targets younger, urban, less-concerned segments matches a smaller and less politically active constituency.

The prescription summary for organisations setting US AI regulatory engagement strategy in 2026:

1. Plan for sustained state-level regulatory activity through 2028 regardless of federal posture; the public demand for regulation drives state legislative output.
2. Operate organisational AI practice above the current regulatory floor; align with public preference and reduce future-regulation exposure.
3. Engage the older-adult, educated, middle-income constituency as the primary regulatory voice in communication and product positioning.
4. Track the federal-state conflict outcomes; the resolution shapes which regulatory layer carries enforcement weight.
5. Treat the deregulatory federal posture as a temporary political position, not a public mandate; plan for the underlying public preference to express through alternative venues (state legislation, civil litigation, consumer pressure, electoral cycles).

> The US public preference for more AI regulation is structural and bipartisan. The federal posture is moving in the opposite direction. The mismatch creates a specific organisational regulatory environment that requires sophisticated engagement. Plans that read the federal posture as the dominant signal will mis-position the organisation for the actual public expectation. Plans that engage with both signals, federal deregulation pressure and underlying public regulatory preference, produce more resilient strategic positioning.

For US-operating organisations setting AI regulatory engagement strategy through 2028, the prescriptive anchor is to plan for regulatory pressure that comes through alternate venues (state, civil, market) even as the federal channel produces deregulatory signal. The public-opinion data anchors this expectation. The organisational practice that gets ahead of likely future regulation rather than minimising current compliance load will be aligned with the underlying public preference. Plans calibrated against current federal posture only will be mis-aligned.

---

### Sources

- **Primary**: Stanford AI Index 2026, Chapter 9 (Public Opinion) 9.3 — [hai.stanford.edu/ai-index/2026](http://hai.stanford.edu/ai-index/2026?ref=aiadoption.org)
- **US state-level public opinion**: [Civic Health and Institutions Project (CHIP50)](https://chip50.org/?ref=aiadoption.org), 2025 — 50-state survey on federal AI regulation expectations
- **Cross-reference**: AI Index 2026 Chapter 8, 8.4 — 150 state AI bills enacted in 2025; December 2025 federal preemption framework