Half your people are faking their AI skills

Your AI rollout dashboard might be lying to you, and your people are the ones writing the fiction.

Visier just released a pulse survey of 1,000 U.S. full-time employees, and 48% say they've exaggerated their AI usage or expertise to colleagues or leadership. Visier has a name for it: Performative AI.

That's not a fringe behavior. That's half the room nodding along in your AI town hall.

The same survey found 45% of employees feel pressure to use AI at work. Seven in 10 are worried about what AI could do to their careers, with job loss topping the list.

Pressure plus fear produces exactly what you'd expect. People perform the version of themselves they think leadership wants to see.

If you're measuring AI adoption by logins and license counts, you're measuring the performance, not the person.

The Evidence

The most telling numbers aren't about AI at all. They're about leadership.

Only 28% of employees say their leaders fully understand how they use AI today. Just 29% believe leaders recognize their concerns about it. Only 33% think leaders understand how much training and support they actually need.

And the plan? 32% say their employer doesn't have a clear plan for how AI affects jobs. Another 27% aren't sure one exists. That's 59% of your workforce operating without a map.

Visier's separate workforce planning research shows the same gap from the other side: 87% of leaders call aligning workforce and business strategy their top business need, but only 29% say their organization is actually good at that kind of planning.

In our own research behind Most Loved Workplace®, covering 2.8 million employees across more than 1,800 organizations, emotional connectedness doesn't decline evenly. In a global study of a multinational manufacturer, it eroded structurally between year one and years three to four of tenure.

That's where AI pressure lands hardest. Your experienced people know exactly what happens to someone who admits they're behind, so they don't.

The upside: when Visier asked what would help, only 12% said nothing would move them. People told leaders what they want. 32% want more AI training, 31% want more transparency from leadership, and 24% want more room to advance.

Look at how a certified Most Loved Workplace® handles it. KnowBe4 runs daily company-wide briefings, a "no-door" policy and "Ask Me Anything" sessions with its CEO, who encourages every employee to spend one hour a day learning new skills. When learning is expected out loud, nobody has to fake it.

The Fallout

• For Employers: Adoption metrics built on self-reporting are now suspect. If half your people overstate their AI use, your rollout ROI, your skills inventory and your workforce plan may all rest on performance rather than reality.

• For Employees: Faking fluency buys time, not safety. The people who say "I don't know how to use this yet" early, in organizations that make that safe, are the ones who'll actually build the skill.

• For Investors: Watch the gap between announced AI adoption and measured capability. Companies that can't see past performative usage will overpay for tools and underinvest in the people expected to use them.

Performative AI is a symptom. The cause is a workforce that doesn't trust leadership to read them accurately, and that's measurable.

Most Loved Workplace® measures emotional connectedness with the Love of Workplace Index (LOWI), validated at a .95 coefficient alpha. Certified organizations see 48% lower turnover. That's the signal you want before it shows up in exit interviews.

👉 Read the full analysis and get your free Opportunity Report