The AI questions your Gen Z employees are actually asking

Most AI conversations start with technology. The more interesting ones start with people.

When I talk to Gen Z employees, four questions come up again and again:

• Am I being replaced or upgraded?
• How do I gain experience if AI does the beginner work?
• What makes a human valuable in an AI-powered workplace?
• Is AI empowering me or monitoring me?

These are culture questions. And most organisations are answering them with technology rollouts.

A 2026 Gallup study, commissioned by the Walton Family Foundation and GSV Ventures, tracked a sharp shift in how young adults feel about AI at work. Excitement dropped 14 percentage points to just 22%. Hopefulness fell 9 points. Anger rose by 9. Nearly half (48%) of young workers now say the risks of AI outweigh the benefits, up from 37% the year before.

That's a significant swing in 12 months.

Researchers at the University of Pennsylvania, including Angela Duckworth, published findings in Harvard Business Review showing that Gen Z's relationship with AI is more pragmatic than personal. They use it constantly (even when told not to), but fewer than 3 in 10 trust AI-assisted work. And virtually none trust work done by AI alone.

The Oliver Wyman Forum's study of 300,000 respondents found 68% of Gen Z workers feel anxious about automation. A Fortune report from May 2026 found 62% of Gen Z workers admitted to treating AI as a crutch, while simultaneously worrying it's making them lazier and less capable of thinking critically.

So here's the picture: a generation that uses AI daily, depends on it, and trusts it less every quarter.

The gap no one's designed for yet

Organisations are investing heavily in AI tools, policies, and training programmes. But very few are experimenting with the cultural shifts needed to help people actually thrive alongside these tools.

Think about the four questions I listed. Every one of them is about identity and trust. They're about what it means to grow in a workplace where the entry-level tasks that used to build competence are increasingly automated. They're about whether your manager sees you as a person developing a career or a node in a system being optimised.

Answers to these questions are not easily solved with a training module.

Stanford researchers have documented a 13% employment decline for 22-to-25-year-olds in AI-exposed roles since late 2022. And more than 40% of CEOs plan to cut junior roles anyway, according to Oliver Wyman. Gen Z can see these numbers. They're paying attention. The anxiety is rational.

What would a culture-first approach look like?

I think the answer is smaller than most people expect.

It starts with asking better questions. Things like: How do we educate line managers to communicate when AI should be used or not? How do we help someone build judgment when AI handles the routine? What does mentoring look like when junior staff skip the grunt work?

Then it means testing those questions. Small experiments. Real teams. Actual data.

That's one of the ideas behind CultureOS Field Labs, something I've been co-designing with Alex Searle. The premise is simple: instead of debating the future, test it. Bring Gen X and Gen Z together. Run disciplined experiments. See what you learn.

Early experiments are looking at things like: What happens when you redesign onboarding around AI-augmented learning instead of AI-replaced tasks? How does transparency about AI monitoring change trust scores within a team? When Gen Z employees co-design the AI policy (rather than receive it), does adoption and confidence shift?

We don't have all the answers yet. That's the point.

Why this matters now

The Gallup data shows sentiment moving fast and in the wrong direction. If organisations wait for the culture problem to become a retention problem, they've already lost time.

Give teams permission to experiment with how humans and AI actually work together, and feed what they learn back into the system. Culture has always been built through practice, through the repeated small choices people make about how to treat each other and how to get things done. AI changes the context for those choices. It doesn't change the fact that culture is where the real work happens.

If you're interested in how we're approaching this at CultureOS, I'd love to hear from you: www.culture-os.com.