10/08/2026
Sarah's company rolled out an AI coaching platform.
It flagged "potential disengagement" on her team and gave her scripted questions for one-on-ones. She asked them. Her team gave polite, surface-level answers. The dashboard showed green checkmarks.
Nothing changed.
Her team didn't lack structured conversations. They lacked the felt sense that it was safe to tell her what was really going on.
That's the gap AI can't close. Psychological safety is a neurobiological response to human connection. It's built when a leader admits what they don't know, responds to mistakes with curiosity, and follows through on small commitments.
And in a culture that's already unsafe, adding AI tools can make it worse: surveillance anxiety, depersonalized support, and a veneer of care that people see through quickly.
This isn't an anti-AI argument. AI is useful for surfacing patterns, providing learning resources, and freeing up leaders' time. But it's a support tool. The foundation has to be human.
In our new post, we break down where AI falls short on each of the S.A.F.E.T.Y.โข drivers (Security, Autonomy, Fairness, Esteem, Trust, and You), the risks of deploying AI in unsafe cultures, and what a human-centered, tech-supported approach looks like.
Read it here: https://brainleadership.com/ai-cant-fix-a-psychologically-unsafe-culture-at-work/