What young voters think about Trump, ICE, Israel, and AI
Plus, new polling on the prevalence of antisemitic beliefs across the American public, the 2028 primaries, presidential powers, and more
The Yale Youth Poll, an undergraduate-led research project at Yale University, today released a new poll of young American registered voters (aged 18-34) and the general registered voter population. The poll sampled 3,429 registered voters, including 2,008 voters aged 18-34, allowing for comparisons of views across generations. Fielding was done by Verasight from 3/9/2026 to 3/23/2026, and results were weighted for age, sex, race, education, and party identification. The survey was conducted in English, and the margin of error after weighting is ±1.4 percentage points for the full sample and ±2.0 percentage points for the youth sample.
The full results can be found on our website, but we wanted to highlight some of the most interesting findings here.
Young Voters Massively Disapprove of Trump, Plan to Vote for Democrats in 2026
Most voters (57%) disapprove of President Donald Trump’s handling of his job in the White House, compared to 41% who approve. Young voters in particular overwhelmingly disapprove of the president’s performance, with large majorities of voters aged 18-22 (68%), 23-29 (72%), and 30-34 (75%) expressing disapproval.
Since the fall 2025 Yale Youth Poll, President Trump’s approval has decreased, particularly among young voters. In fact, Trump’s net approval has fallen among both men and women of every age bucket under 35 years old. Interestingly, women ages 34-44 turned against Trump by the largest margin (28 points), while men in that age group actually swung more in favor of Trump than any other gender-age group (13 points). Note that the fall and spring samples are slightly different and that these are relatively narrow crosstabs, so the specific numbers should be taken with a grain of salt; nonetheless, some of these shifts were large enough to warrant mention.
Although voters disapprove of Trump by large margins, Democrats lead the generic ballot by just two percentage points overall. Note, though, that this survey offered the option of “Don’t know” and did not force undecided voters to pick a side; surveys that do push leaners typically find Democrats doing better in the generic ballot, particularly among respondents who say they are likely to vote.
Still, majorities of young voters are planning to vote for Democrats, including 52% of 18-22 year-olds, 58% of 23-29 year-olds, and 62% of 30-34 year-olds. Unfortunately, the subsample of 18-22-year-olds is relatively male-skewed, so it is hard to determine whether the overall gap between younger and older Gen Z voters is meaningful. In addition, younger voters have a higher share of voters who “don’t know” for which party they plan to vote, so these findings are likely to change as the midterm elections near.
In accordance with our findings from the fall, younger voters are more liberal than older voters, with majorities of those 18-22 (50%), 23-29 (51%), and 30-34 (54%) identifying as on the liberal side. Also consistent with our fall results, the youngest voters appear somewhat more conservative (especially more who identify as “slightly conservative”) and less moderate than other young voters.
When presented with several pairs of issues from a list of 30, voters selected “cost of living/affordability” most often (84%) as more important to their vote, followed by healthcare (75%), democracy (75%), and corruption (72%). Voters ages 18-34 held the same four priorities, with one additional one: housing. Voters care least about Israel and Palestine (28%), Ukraine and Russia (27%), China (27%), artificial intelligence (24%), acceptance of gay people (22%), and acceptance of transgender people (19%), despite the prominence of these issues in the news. Young voters (18-34) prioritized housing, K-12 education, higher education, abortion, climate change, acceptance of gay people, and acceptance of transgender people greater than 10 percentage points more often than the overall population.
Vance Leads 2028 GOP Primary; Democratic Primary Split By Age
Among a list of potential Republican candidates for president in 2028, only two receive more than 10% support. Vice President JD Vance leads the field, with 43% Republican support, while Secretary of State Marco Rubio holds a distant second at 17%. Notably, Republicans aged 18-22 were most likely to write in another candidate not on the given list (24%), compared to just 2% of Republicans ages 65 years and older who did so.
Despite his constitutional ineligibility for a third term, President Trump receives 47% support when included on a list of potential candidates for the 2028 Republican nomination. In this scenario, support for JD Vance and Marco Rubio drops to 18% and 8%, respectively. Republicans ages 18 to 22 were much less likely to back Trump, with only 22% supporting him compared to 47% overall.
When it comes to general election strategy, just over 50% of Republicans believe that their party should focus on energizing its base with “conservative, America-first policies,” while about 30% think the party needs to moderate to appeal to a wider range of voters. About 20% of Republicans see no need to change much from 2024. Support for each Republican strategy remains relatively constant across all age cohorts. Among self-identified “extremely conservative” Republicans, 72% favor the base-energizing strategy and just 8% favor moderation. Among self-identified moderate Republicans, those numbers nearly flip: 51% favor moderation and 29% favor energizing the base.
On the other side, the Democratic presidential primary is wide open, with former Vice President Kamala Harris (20%) and California governor Gavin Newsom (19%) essentially tied at the top, followed by former Transportation Secretary Pete Buttigieg (14%) and New York Representative Alexandria Ocasio-Cortez (13%). Support varies sharply by age: younger voters (ages 18-34) favor Ocasio-Cortez (32%), followed by Harris (20%), while older voters (ages 65+) prefer Newsom (28%) and Buttigieg (19%).
Democratic voters on the whole picked Newsom (72%) and Arizona Senator Mark Kelly (70%) as most likely to win the general election against a Republican. No candidate was as dominant in perceptions of electability among 18-34-year-old Democrats, but Newsom still leads (65%), followed by Harris and Minnesota governor Tim Walz, though this may be an artifact of higher name recognition.
On electoral strategy, Democratic voters are nearly evenly split between prioritizing moderation to pick up swing voters (47%) and energizing the party’s base with progressive policies (45%), with only a small minority (7%) favoring continuity with the party’s 2024 approach. This divide is strongly correlated with age: Younger Democrats prefer a progressive strategy (56% among those ages 18-34), while older Democrats lean toward moderation.
In addition, Black (37%), Hispanic (39%), and female (41%) Democrats are less likely to prefer energizing the base by running on more progressive policies than white (48%), Asian (51%), and male (49%) Democrats. Predictably, the vast majority (73%) of self-identified “extremely liberal” Democrats would like their party to focus on energizing the base, while a majority (56%) of moderate Democrats prefer a strategy of moderation. Unlike the fall 2025 poll, no clear educational divide emerges among Democrats on this question.
Voters Are Largely Skeptical About AI
Young adults (ages 18-34) are the most likely to report using any form of AI. Only 19% reported that they have never interacted with AI, compared to 41% of those age 65+ who said the same. Men were also more likely to report using AI to a greater extent than women, with 21% of men reporting that they use it daily compared to 16% of women.
AI usage is also very correlated with education, with the share of voters who report never using AI decreasing the more educated they become: 41% of those who have a high school or less education say they never use AI, compared to just 15% of those with a graduate or professional degree reporting the same.
AI usage varies widely based on the specific task. Voters are most likely to use AI for everyday tasks, such as making meal plans, creating playlists, or writing emails (46%) and for work or academic purposes (45%). Respondents are far less likely to report talking to AI as they would to a friend (22%) or romantic partner (8%), using AI as a therapist (18%), or asking AI for advice on major life decisions (21%).
The vast majority of voters believe that humans should perform all or most of the work in a variety of situations, including driving public transportation (83%), helping with K-12 learning (87%), filling out personal tax forms (68%), and making decisions regarding military combat (87%), patient care (88%), job hiring (84%), or government benefits eligibility (79%). In driving public transit and making decisions in military combat situations, the majority of voters (58% and 54%, respectively) rejected the use of any AI at all. In every situation polled, women were more likely than men to believe that humans should do all the work without any AI involvement.
Both the youth sample and general population share similar opinions when it comes to the greatest risks posed by AI. Voters are most worried about AI-generated misinformation and about increased reliance on AI reducing independent thinking. However, young voters (18-34) are far more concerned about the environmental impacts of AI than the general population, among whom the environment is the second least concerning risk posed by AI adoption. Voters aged 18-34 place less emphasis than the overall sample on blue-collar worker displacement, sophisticated cyberattacks, and possible AI superintelligence. Across both groups, the issue of plagiarism and intellectual property theft are seen as relatively minor concerns compared to the other risks.
Across both the youth sample and the general population, the top four perceived benefits of AI are consistent: supporting people with disabilities, enabling breakthroughs in medicine and science, performing dangerous jobs, and facilitating real-time language translation. Both groups rank autonomous transportation and increased accessibility to artistic expression as the least important benefits.
Voters Hold Negative Views of ICE, Executive Overreach
We presented voters with seven presidential powers that the Trump administration has asserted, two of which (vetoing bills and issuing executive orders) are clearly constitutional, and the rest of which are more controversial. Americans are largely skeptical that presidents should have these powers.
Majorities of voters disagreed that the president should have the authority to carry out military strikes without congressional approval (53%), exert control over independent financial regulatory agencies (55%), direct the Department of Justice to investigate political opponents (58%), cancel congressionally-approved funding without congressional approval (60%), or continue their administration’s policies in defiance of court rulings (67%), compared to much smaller proportions who agreed.
Still, substantial numbers of voters even opposed giving the president the authority to veto bills (28%) or issue executive orders (43%) — both of which are constitutional; see Article I, § 7 and Article II, § 1 — perhaps as a result of widespread disapproval of the Trump presidency.
More than half of voters surveyed (57%) consider the Supreme Court’s ruling on presidential immunity — Trump v. United States (2024), in which the Court held that presidents have presumptive immunity for all official acts, even after leaving office — as dangerous due to the decision’s implications for abuses of power. About a third (34%) of respondents view the decision as necessary to protect the president’s ability to act without fear of politically motivated persecution. Younger voters are especially opposed to the Court’s presidential immunity ruling, with 75% of voters aged 18-34 viewing it as dangerous.
The poll presented voters with several situations and asked whether each constituted political violence.
Physical acts received the greatest consensus, with 83% of respondents backing the notion that “beating someone up because of their political views” constitutes political violence and 81% believing the same about property destruction.
Views on non-physical actions were more divided: 57% consider trying to get someone fired for their views as political violence, while 45% classify each of ripping down political posters or using racial or religious slurs to insult someone as such.
Rhetorical actions were least likely to be considered political violence. Only 30% described calling political opponents “fascist” or “authoritarian” as political violence, and just 27% said the same about shouting at a political speaker.
Younger respondents (ages 18-34) were generally less likely than older respondents to categorize non-physical actions as political violence.
Voters perceive Immigration and Customs Enforcement (ICE) less favorably than other law enforcement agencies, such as the police. A strong majority (67%) agrees that police make them feel safer, with 37% strongly agreeing. In contrast, attitudes toward ICE are much more negative: only 38% of voters agree that ICE makes them feel safer, while 48% disagree, including 39% who strongly disagree.
Voters are divided about what to do about the ICE budget moving forward. Half of voters (50%) support reducing the ICE budget (23%) or eliminating it entirely (27%), while 42% prefer maintaining (21%) or increasing (21%) it. Younger voters are much more skeptical of ICE, with 68% of voters ages 18-34 hoping to eliminate (44%) or reduce (24%) ICE’s budget.
Young Voters Most Likely to Express Antisemitic Sentiments, Anti-Israel Positions
Building on the work of our fall 2025 poll, we tested three statements commonly considered to be antisemitic, which were inspired by a series of papers by political scientists Eitan Hersh and Laura Royden:
Jews in the United States are more loyal to Israel than to America.
It’s appropriate to boycott Jewish American-owned businesses to protest the war in Gaza.
Jews in the United States have too much power.
Similar to the fall, most voters (67%) agreed with zero of the three statements, though some agreed with one (17%), two (9%), or even all three (6%).
Younger voters were again more likely to agree with these statements: 10% of voters ages 18-34 agreed with all three statements, compared to just 2% of voters ages 65 and above.
A racial gap emerged as well. Black voters were most likely to agree with at least one statement (46%), followed by Asian (39%) and Hispanic (37%) voters, while the same was true of 29% of white voters.
Among the seven ideological groups surveyed, self-identified moderates were the most likely to agree with at least one statement (37%), while little variation emerged between liberals and conservatives. This finding differs from the fall results, which showed no similar ideological trend.
Voters who receive their news from TikTok (41%), Instagram (39%), X (38%), and Reddit (39%) were most likely to agree with at least one of the statements, whereas voters who get their news from television, newspapers, or news websites were least likely to agree with any statements.
We presented respondents with six statements about Israel and the American-Israeli relationship spanning the political spectrum. These statements were paraphrased or taken directly from Nick Fuentes, Ted Cruz, Nikki Haley, Ritchie Torres, Chuck Schumer, and Rashida Tlaib.
The only statement to receive support from a majority of respondents was Schumer’s position (59%) that “Israel is a democracy and the safe haven of the Jewish people, but they have a moral obligation to ensure humanitarian treatment of Palestinian civilians.” While 66% of Democrats strongly or somewhat agreed with this statement, a notable 55% of Republicans agreed as well, making it the most popular position across both parties.
Note that many voters were hesitant to take a position at all: About 30-40% of respondents for each statement selected “Neither agree nor disagree.”
A substantial age gradient emerges with these statements. As the age of the respondents increases, they are less likely to agree with the two strongly anti-Israel statements (those from Fuentes and Tlaib) and more likely to agree with the others. Notably, 54% of voters ages 18-34 believe that “Israel is an apartheid state,” compared to just 34% of voters overall, highlighting a key generational divide.
You can find the full results of our spring 2026 survey on the Yale Youth Poll website.
The Yale Youth Poll is an undergraduate-led research organization. During the 2024-2025 academic year, the Yale Youth Poll received funding from the Democratic Innovations Program at Yale’s Institution for Social and Policy Studies. For the 2025-2026 academic year, the Yale Youth Poll has received funding from the Yale Program for the Study of Antisemitism and programmatic support from Yale’s Institution for Social and Policy Studies.
The Yale Youth Poll operates with complete editorial and methodological independence. All polling activities, including questionnaire development, data collection, analysis, and interpretation of results, are conducted entirely by the undergraduate students behind the Yale Youth Poll. The findings and conclusions presented on this website and on our Substack reflect solely the work and analysis of the Yale Youth Poll and do not constitute an official position of, nor are they endorsed by, the Democratic Innovations program, the ISPS, the Yale Program for the Study of Antisemitism, or Yale University.