§ Claim under review · Adoption
"A new Pew Research Center poll found 52 percent of Americans are 'more concerned than excited about the increased use of AI in daily life,' up from just 37 percent in 2021"
Verdict
Accurate
Confidence
HighSummary
This checks out. Pew Research Center published a survey on August 18, 2026 finding that 52 percent of US adults say they are more concerned than excited about the increased use of AI in daily life, up from 37 percent when Pew first asked in 2021. The quoted phrase is Pew's own question wording, and the survey polled 3,488 US adults from June 22 to 28, 2026 using its probability-based national panel, with a reported margin of error of about 1.8 points. One piece of context the post leaves out is that 52 percent is not a new record. Pew recorded the same 52 percent in 2023 and 50 percent in 2025, so the overall figure has been roughly flat at a high level for three years rather than steadily climbing. What is genuinely new in this wave is that concern among adults under 30 rose to 55 percent and the share who are more excited than concerned fell to a record low of 9 percent. The post's wider argument, that the AI industry's messaging caused this shift and that Americans now want AI gone, is not something the poll measures, and 37 percent of respondents said they were equally concerned and excited.
The readings
key figures from the evidenceUS adults more concerned than excited about AI, 2026 wave
US adults more concerned than excited about AI, 2021 baseline
same figure recorded in Pew's 2023 wave, not a new high
Why this verdict
Evidence
Pew's own publication states the figures the post cites. Today, 52% of Americans say they are more concerned than excited about the increased use of AI in daily life, up from 37% in 2021. Another 9% are more excited than concerned and 37% say they are equally excited and concerned, according to a Pew Research Center survey conducted June 22-28, 2026. The wording quoted in the post is Pew's own question wording, and the 2021 comparison point is the first year Pew asked the question. Pew describes the sample: 3,488 US adults were surveyed from June 22 to 28, 2026, all members of the Center's American Trends Panel, and the survey represents the views of the full US adult population. Reporting on the release gives the margin of sampling error as 1.8 percentage points .
One piece of context the claim omits: 52% is not a new high. Pew's 2023 wave of the same question already found 52% of Americans saying they feel more concerned than excited about the increased use of artificial intelligence , and the intervening 2025 wave found half of respondents saying increased use of AI in daily life makes them feel more concerned than excited, with 38% equally concerned and excited and 10% more excited than concerned . Coverage of the new release described the figure as matching the all-time high in 2023 since the survey began , while the 9% "more excited" share was the lowest recorded .
Findings
✓ What's accurate 6
- The 52% figure is real, current, and correctly attributed to Pew Research Center.
- The quoted phrase is Pew's actual question wording, not a paraphrase.
- The 37% figure for 2021 is correct and is the first year Pew asked this question.
- The poll is genuinely new relative to the post: fielded late June 2026, published Aug 18, 2026, five days before the post.
- The methodology is public and strong: a large probability-based national panel with a stated margin of error.
- The 15-point rise from 2021 is far larger than sampling noise, so the long-run direction of the claim holds.
≈ What's misleading 3
- Omitted qualifier: the claim jumps from 2021 to 2026 and skips the intervening waves. Pew recorded 52% in 2023 and 50% in 2025, so the headline share has been essentially flat, within sampling error, for three years. The "up from just 37%" framing invites a reader to infer a continuing climb when the recent trend on this specific measure is a plateau at a high level. The genuinely new movement in this wave is among adults under 30 and in the "more excited" share falling to a record-low 9%, not in the topline.
- Causal overreach: the post attributes the shift to the AI industry's "out-of-touch outbursts" and "fearmongering," and asserts that the more Americans learn about AI's environmental and labor effects "the more they want it gone." The Pew survey measures sentiment, not its causes, and provides no test of that attribution. The poll cannot carry the causal argument the post builds on it.
- Exaggeration: "they want it gone" overstates what the instrument measures. The largest single non-concerned group, 37%, reported being equally concerned and excited, which is ambivalence rather than rejection.
? What's uncertain 2
- I retrieved Pew's short read and newsletter text but not the full topline PDF and methodology statement for this specific wave, so the exact question order and the full year-by-year series are read from Pew's summary text and prior-wave publications rather than from the topline document itself. The 1.8-point margin of error comes from named-outlet reporting, not from a Pew methodology page I opened.
- The post's broader causal narrative about industry messaging is not something this poll, or any source found, tests. It is uninvestigated as a separate claim.
Sources
6 of 7 linked to recordsPew Research Center short read, "Young US adults are increasingly wary of AI, concerned it will take jobs," published Aug 18, 2026
Pew Research Center short read, "US public concern grows over role of AI in daily life" (2023 wave of the same trend question)
Pew Research Center Weekly roundup, 8.22.26
SSRS notice of the ATP release (SSRS is Pew's fieldwork vendor)
Axios, "Data center backlash is the new existential threat to AI," Aug 19, 2026
The Hill / NewsNation (Nexstar) coverage, Aug 2026
Gizmodo, TechSpot, Fox News coverage, Aug 19-20, 2026