Skip to main content
    Back to The Human Advantage
    Adoption
    6 min read6 February 2026

    Whether people take on the change and make AI part of their work.

    What Is Psychological Safety? Definition, Research, and How to Build It in Your Team

    Psychological safety is the shared belief that a team is safe for speaking up. Here is what it is, what the evidence shows, why it now decides whether people start using AI, and how to build and measure it.

    Gemma George

    Co-Director & Client Service Manager

    Psychological safety is the shared belief that a team is safe for interpersonal risk. People can raise ideas, questions, concerns or mistakes without fear of punishment or humiliation.

    Amy Edmondson introduced the team concept in 1999. Google's Project Aristotle later found it was the most important factor in its effective teams.

    It matters even more as AI arrives. A 2026 study of 2,257 employees found psychological safety reliably predicted whether people adopted AI tools.

    Three common confusions

    Niceness. Safe teams can debate hard. Disagreement is welcomed and nobody is punished for it.

    Comfort. Safety sits alongside high standards. It lets people ask for help when the work gets hard.

    Agreement. Safe teams surface more dissent, and that improves decisions.

    Why it matters

    Learning from errors. In Edmondson's hospital research, better-led teams reported more errors. They were more willing to discuss mistakes, so they could learn from them.

    Performance. Frazier and colleagues' 2017 meta-analysis linked psychological safety to information sharing, learning behaviour, engagement and task performance.

    Better decisions. People share half-formed ideas and minority views. More perspectives are heard before a decision is made.

    AI adoption. Reich and colleagues found safety predicted whether employees started using AI. It did not predict how often they used it afterwards. Starting is the social hurdle.

    Build this capability in your organisation

    Growth Performance designs bespoke programmes that turn insight into measurable behaviour change.

    Explore Adoption

    What damages it

    • Responding to mistakes or questions with sarcasm or public criticism.
    • Ignoring or dismissing input.
    • Blaming the person who raises a concern.
    • Leaving people out of decisions that affect their work.

    How leaders build it

    Go first. Say "I don't know", "I got that wrong" and "What am I missing?"

    Frame work as learning. Ask "What did we learn?" more often than "What went wrong?"

    Ask and listen. Hold back your own answer until others have spoken.

    Thank the messenger. Try "Thank you for raising this. Tell me more." Your first reaction to bad news sets the norm.

    Make AI discussable. Microsoft's 2026 Work Trend Index found 45% of AI users feel it is safer to stick to current goals. Ask openly where AI helped and where it failed.

    How to measure it

    Edmondson's seven-item survey remains the most widely used measure. Adapted items work well as short pulse checks:

    1. If you make a mistake on this team, it is often held against you.
    2. Members of this team are able to bring up problems and tough issues.
    3. People on this team sometimes reject others for being different.
    4. It is safe to take a risk on this team.
    5. It is difficult to ask other members of this team for help.
    6. No one on this team would deliberately act in a way that undermines my efforts.
    7. Working with members of this team, my unique skills and talents are valued and utilised.

    Items 1, 3 and 5 are reverse-scored.

    Try this

    Take the free Team Psychological Safety Audit with your team. Discuss the lowest-scoring item together and agree one change.

    Safety around AI helps people take on the change, which is Adoption, one of the three conditions in our Human Advantage Programme.

    Put It Into Practice

    Free tools and resources to apply what you've just read.

    Team

    Team Psychological Safety Audit

    A research-informed diagnostic to measure the psychological safety of your team, the strongest single predictor of team performance, learning, and innovation.

    Take the Assessment

    Team

    Team Trust Builder

    Assess the quality of trust in your team across three dimensions: vulnerability-based openness, reliability and integrity, and psychological safety.

    Take the Assessment

    References

    1. Duhigg, C. (2016) 'What Google learned from its quest to build the perfect team', *The New York Times Magazine*, 25 February.
    2. Edmondson, A.C. (1996) 'Learning from mistakes is easier said than done: Group and organizational influences on the detection and correction of human error', *Journal of Applied Behavioral Science*, 32(1), pp. 5-28.
    3. Edmondson, A.C. (1999) 'Psychological safety and learning behaviour in work teams', *Administrative Science Quarterly*, 44(2), pp. 350-383.
    4. Edmondson, A.C. (2018) *The Fearless Organization: Creating Psychological Safety in the Workplace for Learning, Innovation, and Growth*. Hoboken, NJ: John Wiley & Sons.
    5. Frazier, M.L. et al. (2017) 'Psychological safety: A meta-analytic review and extension', *Personnel Psychology*, 70(1), pp. 113-165.
    6. Microsoft (2026) *2026 Work Trend Index Annual Report: Agents, human agency, and the opportunity for every organization*. Redmond, WA: Microsoft.
    7. Reich, A., Wolfe, D., Price, M., Choe, A., Kidd, F. and Wagner, H. (2026) 'Safety first: Psychological safety as the key to AI transformation', *arXiv preprint*, arXiv:2602.23279.

    Related programme

    Try the diagnostic

    Related articles

    Want to explore these ideas further?

    Let's discuss how we can help your organisation build the human advantage.

    Start a Conversation