Psychological safety has become one of the most cited ideas in modern management writing, and also one of the most misapplied. The most common misreading turns it into something close to a promise that nothing uncomfortable or costly ever happens on the team, regardless of what someone does or says. That's not what the research behind the term describes, and the gap between the two versions causes real damage in both directions.
What the term actually describes
Psychological safety, as originally researched, is a shared belief that a team is safe for interpersonal risk-taking specifically: safe to ask a question that might sound obvious, admit a mistake, challenge an idea, or raise a concern - without fear of humiliation or retaliation for having spoken up. The safety is about the act of surfacing information. It was never a claim about protection from job performance standards, feedback, or consequences for actual conduct.
Where the confusion causes damage
In one direction: teams that redefine safety as comfort start treating any disagreement, correction, or piece of direct feedback as a violation of the team's culture. Real problems stop getting named because naming them reads as "not psychologically safe" - which is close to the exact opposite of what the concept was meant to produce. A team can end up simultaneously calling itself "safe" and unable to have a single honest conversation about a real issue.
In the other direction: leaders who are skeptical of psychological safety as a concept sometimes point to this exact confusion as a reason to dismiss the whole idea, treating it as inherently in tension with performance standards. That throws away something genuinely well-supported - teams where people feel safe raising concerns catch problems earlier and learn faster than teams where people stay quiet to avoid friction.
What actually coexists
Safety and standards aren't in tension when both are defined correctly, because they operate at different points in the same process:
- Psychological safety governs whether a problem, mistake, or disagreement gets surfaced at all.
- Accountability governs what happens once it has been.
A team can be genuinely safe to disclose a missed deadline in - no humiliation, no punishment for the act of raising it honestly and early - and still have a real, proportional consequence once the facts are established, if the situation calls for one. Those aren't contradictory. A team that only has the first half (safety, no real follow-through) tends to drift. A team that only has the second half (consequences, no safety) tends to learn that hiding problems is safer than reporting them - which is worse for accountability, not better, since problems still happen, they just surface later and with less information attached.
A quick check for which version your team actually has
Ask honestly: when someone on the team raises a mistake, a disagreement, or a piece of bad news, what actually happens to them? If the answer is "nothing bad happens to the person for having raised it, though the underlying issue still gets addressed," that's real psychological safety paired with real accountability. If the answer is closer to "the issue itself quietly gets dropped because raising it again would feel unsafe," that's comfort wearing safety's name - and it's worth naming the difference out loud before it hardens into how the team actually operates.
If that check reveals a real gap, naming it out loud is its own conversation worth preparing for - MBA Lab's Compassionate Accountability Planner walks through rebuilding a team's working agreement once you've named what isn't actually working.