WAR PRESIDENT: THE IRAN TRAP

Introduction

The war the president said would take four to five weeks, and has repeatedly declared won, now stretches on for a half-year. It has claimed the lives of at least 18 US service members, killed thousands of civilians across the Middle East, and its core objectives have gone unmet. The Iranian regime remains in power, the status of its nuclear program is uncertain, and Tehran has weaponized its hold on the Strait of Hormuz. Though a memorandum of understanding struck earlier this summer was meant to provide a framework for negotiating a final deal, it was tenuously held and formally expired on August 17.¹

Fighting resumed after a month-long lull, and US economic pressure has only intensified — all of which could be taken as signs that Donald Trump has found himself trapped at home and abroad by a war that no one is winning and that has no clear off-ramp. Trump’s approval ratings have hit record lows, and in our polling, most Americans say it was a mistake to attack Iran, only a third of the country approves of his handling of the war, and three in four Independents now disapprove.² Republicans still back him, but with less conviction than in the spring. The war’s effect on affordability has been immediate and broadly felt. Traffic through the Strait of Hormuz, which before the war carried a fifth of the world’s oil, has slowed to a trickle and oil prices have stayed high. Three months before the midterms, eight in ten Americans — majorities across every political party — say the war has raised the cost of living.

The war may also leave a geopolitical legacy far different from the one the administration intended. More Americans say the likelihood of an Iranian bomb has increased over the past year than say it has decreased. America’s stock of missile interceptors, needed for contingencies in Asia and elsewhere, has run low. And the war has opened questions about the future of warfare and America’s conduct abroad that the public is only beginning to reckon with. Artificial intelligence has been used to identify and prioritize military targets — something fewer than one in ten Americans have heard much about. Strikes on schools and civilian infrastructure, whether intentional, accidental, or merely threatened, implicate America’s moral standing — a majority call deliberate strikes on civilian power, fuel, and water systems unacceptable.

The Trump administration and the American public will be stuck with this war for months and years to come, regardless of when it ends.³

Executive summary

The Institute for Global Affairs (IGA) at Eurasia Group surveyed 1,000 voting-age Americans between August 11 and 14, 2026. The survey, the second in a series of IGA polls tracking public opinion ahead of the 2026 midterm elections, finds a public wary of a war that many believe the United States is not winning, unlikely to achieve its stated objectives, and not worth the costs. Dig beneath the toplines, however, and the war divides Democrats and Republicans — each counted here with the Independents who lean their way — even as President Trump’s political base hedges.

Republicans still back the war and largely approve of President Trump’s handling of it, but with less intensity than in the spring.

Republican strong approval collapsed 15 points since April, and Independents’ disapproval jumped 13 points, though the Republicans’ overall approval has held steady at 74% — virtually unchanged from the spring (73%).

Roughly a sixth of Republicans think the United States made a mistake in attacking Iran (17%), while two-thirds think the United States didn’t make a mistake (65%).

Half of Republicans think the war is making the United States safer (49%), a quarter say it’s making the country neither more nor less safe (26%), and 16% say less safe — fewer than one in ten aren’t sure (8%).

Most Republicans trust the Trump administration to provide accurate information about the war (69%), but about a quarter have doubts (26%) — 17% say they don’t trust the administration much and 9% say not at all.

Nationally, few Americans think that the war has been a quick victory for the United States.

Approval for Trump’s handling of the Iran war hovers around one in three (34%). Disapproval for the president’s handling of the conflict is intense — more than three times as many “strongly disapprove” (46%) as only “somewhat disapprove” (14%).

Roughly as many say the war has made the United States safer (23%) as say it has made no difference at all (22%) — and a plurality think the war is making the United States less safe (46%).

A plurality of Americans — 43% — say Iran is more likely to develop a nuclear weapon today than it was a year ago.

Americans are split on which country currently has the advantage. Slightly more than one-third (37%) say the United States, one-quarter says Iran (25%), and few say neither (14%).

Despite widespread disapproval of the war and low trust in the Trump administration, Americans think the United States is more likely than Iran to honor the terms of a ceasefire. More than twice as many report a great deal or a fair amount of confidence in the United States (53%) than Iran (22%).

Beneath the surface, Democrats and Republicans register very different opinions.

Only 5% of Democrats approve of the administration’s handling of the war, a nearly 70-point gap with Republicans.

Majorities of Democrats (81%) and Independents (59%) think the United States made a mistake in attacking Iran. Seventeen percent of Republicans agree.

When presented with six goals that could justify military action against Iran, half of Democrats say none of them do (51%) — seven times more than the share of Republicans who think the same (7%).

More than seven times as many Democrats blame the United States for the war (53%) as blame Iran (7%). Meanwhile, more than three times as many Republicans hold Iran more responsible (55%) than the United States (15%).

Americans have few shared sources of information about the war, and this shapes their understanding of the US military operation in Iran — its toll on US troops, its use of artificial intelligence, and the harm done to civilians.

A plurality (39%) of Democrats reported hearing “a lot” about a missile strike on a girls’ elementary school in Iran on February 28, 2026. Only 11% of Republicans reported hearing “a lot,” and a plurality (40%) had heard “nothing at all.”

Awareness of US troops killed in the war did not vary as much by party. Roughly two-thirds of Democrats (64%) and Republicans (65%), and six in ten Independents (59%), reported at least some awareness of it.

Independents were the least aware of the impact of US military strikes on Iran’s nuclear facilities: more than twice as many heard “nothing at all” (24%) as “a lot” (10%).

On the military’s use of AI in the war, only 8% of Americans have heard “a lot” and 42% have heard nothing at all — by far the least familiar of the topics we asked about. That holds across party: only 11% of Democrats, 5% of Independents, and 6% of Republicans have heard “a lot.”

Paying attention to public affairs doesn’t close the AI gap. Among Americans who follow government and public affairs most of the time, just 11% have heard a lot about the military’s use of AI — the only topic we asked about on which even the most attentive are more likely to have heard nothing at all (39%) than a lot (11%).

---

Endnotes

1. Institute for Global Affairs is grateful to the several journalists and think-tank experts who advised on the survey instrument. Thank you to IGA CEO Mark Hannah and development and operations associate Sasha Benke. Thank you to IGA interns Ria Mueller and Ammara Elsevier.

2. Jason Lange. “Trump Approval Falls to 33%, Lowest of His Presidency, Reuters/Ipsos Poll Finds.” Reuters, August 17, 2026. https://www.reuters.com/world/us/trump-approval-falls-33-lowest-his-presidency-reutersipsos-poll-finds-2026-08-17/; Fernando Cervantes Jr. “Trump Approval Rating Poll.” USA TODAY, September 8, 2026. https://www.usatoday.com/story/news/politics/2026/09/08/trump-approval-rating-poll/91662625007/.

3. Throughout this report, Democrats and Republicans include Independents who say they lean toward that party, and Independents are those who lean toward neither. Party identification is measured on a seven-point scale; respondents who answered “not sure” are included in the overall figures but not in the party breakdowns. This is the same approach used in our April 2026 report.