Younger Workers Feel Pressure to Check AI Before Admitting Uncertainty
A Use.AI survey reports many workers, especially Gen Z, feel they must consult AI before asking colleagues a question.
According to TechRadar, a survey from Use.AI of more than 10,000 people found that many workers feel uneasy saying they do not know something at work. The report says the pressure is most acute among younger employees.
TechRadar reports that 71 percent of Gen Z employees said they feel pressure to check an AI tool before raising a question with colleagues. By comparison, 44 percent of workers aged 45 and older said the same.
The survey also found that 56 percent of employees now regard saying "I don't know" as a risk, a change the report links to the arrival of AI tools in the workplace. Almost half of respondents said that checking AI first is simply expected.
Taken together, the findings suggest that AI is becoming a first stop before people turn to coworkers, and that admitting a gap in knowledge is increasingly seen as costly. TechRadar presents these results as survey findings rather than measured changes in workplace behavior.
Background
Admitting uncertainty has long been part of how people learn on the job. Organizational researchers have studied the idea of psychological safety, the belief that a person can speak up, ask questions or admit a mistake without being punished or embarrassed. Teams where people feel safe to say "I don't know" tend to share problems earlier and correct errors faster, while teams where people hide gaps often discover problems late.
Generative AI tools became widely available to the public in the early 2020s and quickly moved into office software, search engines and internal company systems. Many employers encourage staff to use these tools for drafting, summarizing and research. The tools can be fast and convenient, but they can also produce confident-sounding answers that are wrong or out of date, which means users must still verify what they are told.
Younger workers often entered their careers during the same period that these tools spread, so they may have learned habits of quick self-checking early. Managers, meanwhile, may not have set clear expectations about when an AI check is appropriate and when a direct question to a colleague is more useful. Without clear norms, individuals may assume that asking a person is a burden and that asking a machine is the safer default.
Why it matters
The story touches on something larger than one survey. When people feel they must consult a tool before asking a human, a few things can change at once. Knowledge can become less shared, since mentors and colleagues may have fewer chances to explain their reasoning. Newer staff can lose the informal training that comes from asking questions and watching how experienced people solve problems.
There is also a question of accuracy. If workers rely on AI answers to avoid looking uninformed, errors may pass through a team without anyone reviewing them. The pressure described in the survey could push people toward accepting a first answer rather than testing it against experience or evidence.
Finally, the findings speak to workplace culture. A company that wants to benefit from AI tools may also need to protect the habit of honest uncertainty. Leaders who model admitting what they do not know may do more to reduce the pressure than any single tool or policy.
What to watch
Key questions remain open. It is not yet clear how the sample was designed or how representative it is of workers across different industries and countries. Readers should also watch whether follow-up research measures actual behavior, such as how often people check AI before asking a colleague, rather than only how much pressure they report feeling.
Watch for whether employers respond with written guidance on appropriate AI use, and whether that guidance encourages people to ask questions directly. It will also be worth tracking whether teams see fewer repeated mistakes or more knowledge sharing as AI use grows.
Viewpoints
Commentary from our columns: three angles on this story.
The User what changes for people who use it
For most people, this is not an abstract debate. It is a small daily decision: send a quick chat message to a coworker, or open an AI tool and hope the answer is good enough. The survey suggests many people are choosing the tool, sometimes because they feel that asking a person is a bother. That feeling is worth taking seriously. Nobody wants to be the one who keeps interrupting. But a question to a colleague does more than fill a gap. It builds a relationship, reveals context a tool cannot see and lets the person answering learn what the team is actually stuck on. If the tool becomes the default, those small exchanges quietly disappear. Workers may find they know more answers but have fewer people to ask. The practical takeaway is simple: use AI when it helps, but do not let it replace the habit of saying "I'm not sure, can you help me understand this?" That sentence is still one of the most useful things a person can say at work.
The Ledger the business angle
From a business perspective, the number that stands out is the one about risk. When more than half of employees feel that admitting ignorance carries a cost, that is a management signal, not just a personal anxiety. Companies invest heavily in AI tools on the assumption that they will speed up work and reduce friction. But if the tools also make staff reluctant to flag problems, the savings could be offset by slower error detection, duplicated effort and weaker onboarding of newer employees. Those costs rarely show up on a software invoice. They appear later, in missed deadlines or preventable mistakes. Leaders who measure only tool adoption may miss the cultural side of the return on investment. The smarter question is whether the organization gets better decisions and faster learning. A workforce that feels safe saying "I don't know" is often more productive over time than one that quietly checks a chatbot first and says nothing.
The Skeptic open questions and weak points
The headline finding deserves some caution before anyone builds a policy around it. Survey results depend heavily on how questions are worded and who answers them. Asking whether people feel pressure to check AI first is different from measuring whether they actually do it, and feelings of pressure do not always match behavior. The gap between age groups is striking, but it is not clear what drives it. Younger workers may simply be more comfortable with AI tools, or they may be newer and more anxious about appearing inexperienced at any stage of a career. The report also does not settle whether AI is the cause of the change. Workplace norms about asking questions were already shifting for many reasons. Before concluding that machines are making people timid, it would help to see comparable data from earlier years and from organizations with different AI policies. Until then, the result is a useful warning, not a settled explanation.