Half Your Team Is Thriving on AI. The Other Half Is Scared.

The fresh sentiment data on how workers and the public actually feel about AI in 2026 is not one story. It's two, and they're pulling in opposite directions inside your own team right now. One group is energized. One group is scared. And the split does not fall where you'd guess.

If you're setting adoption pace and writing the internal memo about it, you're working with old assumptions. Let me catch you up.

Your team is not one audience

The annual Tech Worker Sentiment Survey, run by Noam Segal and now in its second year, found the tech workforce split almost exactly in half. One half is thriving on AI. The other half is shaken. The survey sorts people into four types: the Energized, the Conflicted, the Disoriented, and the Resentful.

That matters because you probably write one all-hands message about AI and send it to everyone. Half your people read it as good news. Half read it as a threat. Same words, opposite reaction.

So stop treating adoption as a single rollout. The Energized will run ahead no matter what you do. The Disoriented and Resentful need a different conversation, and it's not a louder version of the same pitch.

Burnout jumped, and the top fear isn't layoffs

Burnout climbed 11 points in a single year. That's a big move for one year, and it's happening while you're asking people to learn new tools on top of their real jobs.

The number one fear in tech right now is not losing your job to AI. It's something else, and that changes your comms. If you keep reassuring people that their jobs are safe, you may be answering a question they aren't asking, while ignoring the one keeping them up at night.

Segal also found that managers are the single biggest lever for well-being. Not the tools, not the policy, the manager. So your AI adoption plan is really a management plan. If your frontline managers can't hold a straight conversation about what's changing and why, no rollout deck saves you.

Trust splits by who has something to lose

Zoom out from your team and the same pattern shows up in the public. Oren Etzioni's read of the research lays out the gaps. In China, nearly nine in ten people say they trust AI. In the U.S., barely a third do. Where economies grow fast, AI reads as a ladder up. Where they're mature, it reads as a threat.

Inside the U.S., men are about twice as likely as women to expect AI to be good for society, per Pew. And it's not a usage gap anymore. Women use chatbots as much as men now and still trust them less.

The under-30s are the heaviest users and the most convinced AI will be bad for society. Etzioni's line explains why: the young are the most exposed, since AI is coming first for the entry-level jobs they're trying to land. Optimism tracks who stands to gain. Doubt tracks who stands to lose.

The builders are not the buyers

Here's the split that should make you pause before you copy a lab's playbook. Among AI researchers Pew surveyed, most expect the technology to help the country over the next two decades. Among the public, fewer than one in five do.

Etzioni puts it plainly: AI is being built by the people most enthusiastic about it, for a public that is not. The people designing these systems have their careers riding on the upside. The people answering phones or driving trucks mostly see the risk.

Your vendor's demo comes from the optimistic end of that gap. Your team sits closer to the public end. Don't mistake the builder's confidence for how your own people feel.

The deep cut

Watch the gap between what leaders say and what they do. Xbox CEO Asha Sharma was named to a Federal Reserve task force on AI and jobs three days after announcing 3,200 gaming layoffs, about 20% of the division. The gaming press tore into her, and the timing wasn't even her call. But the lesson holds for you: your team reads your actions against your words, and if the two don't match, they trust neither.

So when you announce an AI push, your people are checking whether you're cutting headcount in the same breath. If you say "AI makes us all more productive" and then quietly trim a team, the Resentful group grows, burnout climbs, and your next rollout lands even harder. Sequence and honesty beat spin. Say what's actually changing before someone else fills in the blank.

Three questions for your team

• Which of the four types are your key people, and are you sending one AI message to a room that's split in half?
• Can your frontline managers hold an honest AI conversation, or are you leaning on all-hands emails to do a manager's job?
• If you announced your AI plan and your headcount plan side by side tomorrow, would they tell the same story?