Employees fear blame over AI mistakes, survey finds

GoTo has published survey findings showing that 83% of employees fear they could be blamed or fired over an AI-driven mistake. The study polled 2,500 employees and IT leaders across several countries.

The data points to a widening gap between the productivity gains workers say they get from AI and their confidence in using it safely at work. Employees who use AI said they save an average of 2.3 hours a day, yet 31% said they had hesitated to use it for fear of the consequences of an error.

The findings also suggest concern over AI mistakes is not theoretical. A quarter of IT leaders said AI had already made mistakes that negatively affected customers or clients, while 23% said the technology had caused financial harm to their company.

Nearly all IT leaders surveyed said they were worried about future incidents. The report found that 91% were concerned AI would make a mistake that affects their organisation.

Policy gaps

One of the clearest divides in the research was between larger companies and smaller businesses on formal AI rules. Only 24% of small businesses and 36% of midsize companies said they had an AI policy in place, compared with 53% of enterprises.

That leaves many workers using the tools without formal guidance on what is allowed, what should be checked, and who is responsible when something goes wrong. The survey also found that 84% of workers felt their employer could do more to encourage responsible AI use.

In practice, many respondents admitted to weak oversight of AI-generated work. More than half, 55%, said they do not always review or double-check AI outputs, even for high-stakes tasks. Another 43% said they had used AI outputs even though they believed they were low quality or suspected they contained errors or fabricated information.

The survey linked some of that behaviour to pressure inside organisations. Sixty per cent of employees said they feel pressure to use AI to increase productivity, while 47% said AI use is considered during performance reviews.

That pressure appears to shape how some workers judge risk. The report found that 42% would rather risk making errors with AI than look "slow" or "replaceable".

Blame and silence

The research also highlighted how AI can complicate accountability in the workplace. More than a quarter of employees, 26%, said using AI allows them to avoid personal accountability at work. A further 25% said it feels safer to blame AI than to admit they made an error, and 17% said they had already blamed AI for a mistake they made at work.

Some employees said concerns about AI errors were not always welcomed by managers. The survey found that 31% believed there is unspoken pressure at work to trust AI and stay quiet about its mistakes.

It also found that 14% had reported AI errors to a manager or leader but were told to stay quiet about them, suggesting some companies may be struggling not only with oversight of AI systems but also with internal reporting culture when problems emerge.

GoTo conducted the study with Workplace Intelligence among 1,250 full-time knowledge workers and 1,250 IT decision-makers. Respondents were drawn from the US, Canada, the UK, Ireland, Germany, Austria, Switzerland, India, Mexico, and Brazil.

The study also examined the spread of agentic AI, which is being adopted by smaller businesses despite lower levels of readiness. It found that 46% of small businesses are already using or piloting agentic AI, compared with 70% of enterprises.

Readiness levels differed sharply. Among small and midsize business IT leaders, 38% said they felt unprepared to manage agentic AI securely, compared with 14% at enterprise companies.

The findings point to a tension many employers now face: AI tools can save time, but their use is outpacing governance and training in many workplaces. The report suggested that as AI takes on more routine work, employees increasingly see creative thinking and adaptability as the skills that will matter most.

More than half of employees, 52%, identified those human skills as the most important to maintain. The result indicates that while automation is reshaping day-to-day work, workers still expect judgement and flexibility to remain central to their roles.

"Our research captures an inflection point for businesses," said Rich Veldran, Chief Executive Officer of GoTo. "In the US alone, there's an opportunity for more than $2.9 trillion* in potential efficiency gains to be captured from effective AI use. But without clear guardrails, that potential is being eroded by anxiety, misuse, and unaddressed risk. The organisations that close the gap with clear accountability structures, training, and policies that give employees confidence won't just speed adoption, they'll be the ones who pull ahead."

Dan Schawbel, Managing Partner at Workplace Intelligence, said workplace pressure around AI was creating the conditions for careless use.

"We're seeing a new kind of workplace pressure: employees feel they must use AI to avoid looking replaceable, but fear repercussions if it goes wrong. That combination is exactly how you get a workforce that uses AI recklessly," he said. "Many organisations are accelerating AI adoption without providing the training, policies, and support employees need to use these tools effectively. The companies that close this gap will be best positioned to unlock AI's full capabilities while building a more productive and confident workforce."