Last updated: August 17, 2026 · By Vishal Swami, Founder & Lead AI Reviewer, AISagely
Young people hate AI CEOs, and a new poll puts a number on it: most young US adults say they don't trust a single sitting AI executive to act responsibly. The survey, run by CNBC's Generation Labs among more than 1,000 US adults ages 18 to 34, asked who respondents trust to handle AI responsibly — and the "don't trust" numbers were ugly across the board.
Short answer: A 2026 CNBC Generation Labs poll of over 1,000 US adults aged 18–34 found deep distrust of AI executives. Palantir's Alex Karp scored worst at 81% "don't trust," followed by Peter Thiel at 79%. Microsoft's Satya Nadella fared best, but still only pulled a 35% trust score. Nearly half of respondents fear AI will hurt their careers.
I test AI tools for a living, which means I spend most weeks reading claims from the same executives this poll asked about. In my testing of the products these companies ship, the gap between the confidence in a CEO's keynote and what the tool actually does when you push on it shows up constantly. That gap is a big part of why I wasn't surprised by these numbers, even though the size of them still is.
What the poll actually found
Futurism reported on August 16, 2026 on the CNBC Generation Labs survey, which asked respondents whether they trust each of nine executives overseeing major AI companies to act responsibly on AI. The headline number: a majority of young adults said "don't trust" for every single name on the list. Nobody cleared a positive trust rating.
Palantir's Alex Karp came out worst, with 81% of respondents saying they don't trust him. Peter Thiel was close behind at 79%. Meta's Mark Zuckerberg landed at 71% don't-trust, Tesla and xAI's Elon Musk at 70%, and OpenAI's Sam Altman at 69%. Microsoft's Satya Nadella scored the best of the nine — and even that was a 35% trust rating, not a majority.
The distrust isn't only aimed at the people. Nearly half of respondents, 45%, said they expect AI to hurt their careers, against just 10% who think it will help. Forty percent said the US government should step in and regulate AI, and 60% said they'd like to see data center construction slow down. One detail from the poll stuck with me: respondents rated the AI technology and the data centers powering it more favorably than the executives running the companies behind them. People aren't rejecting AI outright. They're rejecting the people selling it to them.
Why young people don't trust AI CEOs
Some of this is timing. Layoff announcements keep landing next to AI-adoption announcements, and 18-to-34-year-olds are the cohort most likely to be job-hunting into that headwind. A trust gap that size doesn't come from one bad news cycle, though — it comes from a pattern of claims that don't match lived experience.
You can see that pattern in how AI executives have responded to the backlash itself. Nvidia CEO Jensen Huang, on the No Priors podcast in January 2026, said the negative narrative around AI has been "extremely hurtful, frankly, and I think we've done a lot of damage," pinning the blame on "doomers," according to Gizmodo’s report on the appearance. That's a common move: treat skepticism as a messaging problem rather than a response to what people are actually experiencing — layoffs, chatbots that confidently get things wrong, and pricing that keeps climbing. Telling your most skeptical audience that their skepticism is "hurtful" to you is not usually how you win them back.
Step-by-step: how to decide who to trust on AI claims
You don't have to pick a side between "AI CEOs are all lying" and "AI CEOs are all telling the truth." Here's the process I actually use when a company founder makes a big claim about their own AI product.
1. Separate the claim about the technology from the claim about the company
"Our model can do X" and "you should trust us with your data because of X" are two different claims that need two different kinds of evidence. A capable model doesn't automatically mean a trustworthy company.
2. Find the primary source, not the recap
Press releases and keynote clips are marketing. Look for the benchmark report, the pricing page, or the filing the claim is based on. If you can't find one, treat the claim as unverified.
3. Test the product yourself before trusting the pitch
I don't take a vendor's word for what a model can do — I run it. My AI tool reviews explain the testing process I use, and my AI tool ratings guide covers how to read a review without getting fooled by cherry-picked demos.
4. Check who's disputing the claim, and why
When Anthropic's Dario Amodei and Nvidia's Jensen Huang publicly disagree on how many jobs AI will displace, that disagreement is information. My breakdown of what’s actually happening to jobs walks through how to weigh competing claims like that instead of picking whichever one confirms what you already believed.
5. Watch what the company does with its own money
Spending patterns tell you more than keynotes do. My look at Nvidia’s $750B in circular AI deals is a good example of a story where following the money mattered more than following the press release.
Example prompts you can copy
Use these with any chatbot to stress-test a claim before you repeat it:
- "Find the primary source for this claim: [paste the claim]. Tell me if it's from a company press release, an independent study, or a news report, and give me the direct link."
- "List the specific, testable claims in this statement, separate from the opinions and predictions mixed in with them."
- "Who has publicly disputed this claim, and what's their reasoning?"
These work because they force the model to separate sourced fact from spin instead of just summarizing the confident version back to you.
Common mistakes to avoid
The biggest mistake I see is treating "an executive said it in a keynote" as equivalent to "it's been verified." A confident delivery isn't evidence. The second is swinging the other way and assuming every AI claim is hype — that costs you real productivity gains from tools that do work as advertised. When I tested products from several of the companies in this poll, I found real gaps between the marketing and the output, but also real capability that gets lost when the conversation turns into blanket distrust. Third, don't confuse distrust of a CEO with distrust of the underlying technology — the poll itself shows people rate the two differently, and conflating them leads to worse decisions either way.
The trust gap, by the numbers
| Executive | Company | Don't trust | Notes |
|---|---|---|---|
| Alex Karp | Palantir | 81% | Worst score in the poll |
| Peter Thiel | Palantir co-founder | 79% | |
| Mark Zuckerberg | Meta | 71% | |
| Elon Musk | Tesla / xAI | 70% | |
| Sam Altman | OpenAI | 69% | |
| Satya Nadella | Microsoft | 35% trust | Best score, still a minority |
Source: CNBC Generation Labs poll of 1,000+ US adults aged 18–34, reported by Futurism, August 16, 2026.
Tools that make this easier
If the takeaway here is "verify before you trust," the fastest way to do that is to test the products yourself instead of relying on a keynote. My AI tool reviews page explains how I actually test these products, and Best AI Models (2026) ranks the models I'd currently recommend based on that testing, not on marketing claims. For the career-anxiety piece of this poll — the 45% who expect AI to hurt their jobs — my article on the AI jobs apocalypse is a more grounded look at what's actually happening versus what's being predicted.
My take
The poll numbers are bad for a reason: young adults have been asked to trust a group of executives whose public statements don't reliably match what happens when you actually use their products or watch where their money goes. That's not irrational distrust — it's pattern recognition. I don't think the fix is louder reassurance from the people already being disbelieved. It's more people testing the actual claims themselves, the same way I test a model's marketing page against what it does when you sit down and use it.
Frequently Asked Questions
Do young people actually trust any AI CEO?
Not by much. In the 2026 CNBC Generation Labs poll, every one of the nine executives asked about had a majority "don't trust" score. Microsoft's Satya Nadella came closest to positive, with a 35% trust rating — still short of a majority.
Why do young people distrust AI CEOs more than AI itself?
The same poll found respondents rated AI technology and data centers more favorably than the executives running the companies behind them. The distrust tracks the people and their public statements, not necessarily the tools.
Is the AI jobs fear from the poll realistic?
It's mixed. Nearly half of respondents expect AI to hurt their careers, but the picture is more nuanced than a straight replacement story — see my breakdown of what’s actually happening to jobs for the fuller context.
How can I check an AI CEO's claim myself instead of just trusting or dismissing it?
Trace it to a primary source, separate the technology claim from the company claim, and test the product directly rather than taking a keynote's word for it. The step-by-step section above walks through the process I use.