For the past few years, most workplace discussions about AI have centered on productivity: Where can automation save time, reduce repetitive work, and help employees move faster? Most of us have seen at least some of the answers to those questions. But, as becomes increasingly embedded in day-to-day work, a competing factor is emerging, which is that many employees are not experiencing this shift as a straightforward upgrade, but a source of uncertainty about fairness, expectations, job security, and their long-term prospects.
MetLife’s latest Employee Benefit Trends Study suggests AI’s impact on work is raising the stakes for skills development, clearer management communication, and more robust employee support. The numbers tell the story:
- 80% of employers say AI tools are now part of routine tasks
- 83% of employers AI is helping staff work faster
- 61% of employees worry about AI’s ethical and safety risks
- 59% of employees fear it will make skills obsolete faster than new opportunities emerge
- 24% of employees feel they are competing with AI at their own workplace.
There’s a distinct difference in perspective among employers and employees. The question isn’t whether companies can deploy AI effectively – that’s pretty much a given. It's whether they can do it in a way that creates a positive employee perception, a factor that is much more important to long-term success and growth than short-term results.
Productivity is rising, trust is not
While it’s not surprising that employees have concerns, it’s more interesting that some two-thirds of employers believe AI is creating new friction or mistrust between employees and management, and 54% say they are struggling to adapt to the new ways of working employees now expect.
For sure, AI can improve output, but it can also destabilize workplace culture. If employees do not understand how AI is being used, how their performance is being evaluated, or how their roles may change, productivity gains can happen hand-in-hand with an erosion of trust.
For example, while 91% of employers say employee contributions are valued and fairly rewarded, only 65% of employees agree. In addition, 55% of workers believe success is judged primarily by output, suggesting many feel the workplace is becoming more transactional.
To be clear, MetLife is a benefits provider, not a technology company, so it’s understandable it has a slightly different take on AI momentum and why it looks at AI as a workforce support issue. If employees are under pressure to reskill continuously while also worrying about displacement, ethical misuse, or unrealistic productivity targets, the consequences will show up in engagement, mental health, and retention.
That’s not to say the company is opposed to the use of AI – it doesn’t suggest that at all. Rather, it sees an opportunity for a “Success Reset” – brings “human skills, connection and modern benefits together to help organizations move from uncertainty to confidence as work continues to rapidly evolve.” Basically, MetLife argues that companies need to rethink how they define achievement and support in an AI-enabled workplace, that traditional metrics, like speed, output, and efficiency, are increasingly easy for AI to optimize. That can leave workers feeling the goalposts have shifted without warning. It’s so-called reset means broadening the definition of success to include adaptability, collaboration, and judgment – qualities that AI does not replicate.
The study finds that holistic well-being continues to decline, and that connection, skill development, and meaningful benefits are becoming more important as technological change accelerates. This aligns with broader Federal policy: Earlier this year, the U.S. Department of Labor released an AI literacy framework encouraging employers and education systems to expand AI training and readiness programs. The recommendation reflects a growing recognition that adoption can’t be separated from worker preparedness for maximum success.
What does this look like in practice? It might mean giving employees clear explanations of how AI tools are being used and what role human judgment still plays. It might mean investing in reskilling before obsolescence becomes acute, rather than after. It could also mean offering benefits that help absorb the anxiety that accompanies rapid workplace change.
Adoption vs. Navigation
The organizations most likely to benefit from AI may not be the ones that automate the fastest. While it may seem that way initially, in the longer term, success may be defined by how well companies reduce ambiguity for their workforces. By and large, workers are not rejecting AI. Rather, they are asking for clarity around how it will be used, how they will be supported, what skills they need next, and whether success will continue to be defined in human terms rather than purely algorithmic ones.
When AI adoption is framed purely as a performance initiative, companies run the risk of deepening the kind of mistrust the MetLife brings to light. In that scenario, result is surface-level compliance without genuine confidence, which likely won’t serve organizations well in the longer term.
That points to the real divide, which is not between companies that AI and those that do not. Instead, it’s between those that treat AI as a technology employees must keep up with, and those that treat it as a change employees need help navigating. Looks at it this way: AI is less like figuring out the your new laptop, phone, or CRM application, and more like the challenge many employees faced when they were forced to an overnight transition to remote work during the pandemic. Those who had the infrastructure and support to adapt succeeded to a much greater degree than those who didn’t.
AI is here, it’s not going away, and it has the potential to drive profound change. How and to what extent that happens is up to each company and its leadership. Companies that view AI as more than merely a technology issue are likely to see better results. They will see greater trust and buy-in and, as a result, higher loyalty and retention and longer-term productivity and business results.
Edited by
Erik Linask
