Published on
If you are a boss using artificial intelligence (AI) to communicate with your employees, beware: your professional credibility may be at stake, a new study has found.
AI is rapidly making its way into the workplace. In the European Union, 13.5 per cent of businesses with at least 10 employees reported using AI last year, compared to 8 per cent in 2023, according to Eurostat. And a 2024 survey by Microsoft and LinkedIn found that 75 per cent of knowledge professionals around the world used generative AI at work.
While this shift can help boost workers’ productivity, it can also come with downsides – for example, hurting relationships between colleagues.
For the new study, which was published in the International Journal of Business Communication, researchers surveyed more than 1,000 full-time professionals in the United States to understand how they perceived emails written with low, medium, or high levels of AI assistance.
They were given scenarios where they were randomly shown one email, which they were told was written either by themselves or by their supervisor, and asked to rate it on traits such as professionalism, effectiveness, sincerity, and how caring it was.
While AI-assisted messages were generally considered efficient, effective, and professional, the study highlighted a perception gap in messages written by employees and managers.
“Overall, professionals view their own AI use leniently, yet they are more sceptical of the same levels of assistance when used by supervisors,” Anthony Coman, a researcher at the University of Florida and one of the study’s authors, said in a statement.
In other words, people tend to judge AI use more strictly when it comes from their manager than when they use it themselves.
This difference was especially significant if a supervisor’s message relied more heavily on AI, going beyond simple grammar, proofreading, or editing. Only 40 per cent of employees viewed supervisors as sincere when they used high levels of AI, compared to 83 per cent for low-assistance messages, the study found.
“In some cases, AI-assisted writing can undermine perceptions of traits linked to a supervisor’s trustworthiness,” Coman said.
However, staffers’ perceptions also appeared to vary depending on the purpose of the message. If the email was seen as purely informative, employees tended to view AI use positively. But if it was perceived as relationship-based or motivational, they were far less accepting.
The study has some limitations. The researchers drew conclusions based on one hypothetical scenario, and people’s perceptions could also be biased by the power dynamic between employee and supervisor.
Even so, the results are in line with other findings. Researchers have pointed out how the use of AI in professional settings can hinder one’s reputation, while others recently found that people who admit they use AI at work can lose their colleagues’ trust.
