MIT Tech Review
AI agents are not your “coworkers”
agentic
What happened
A study by Boston University professor Emma Wiles found that managers caught 18% fewer errors when work was presented as coming from an agentic 'AI employee' rather than a standard chatbot. The research warns against anthropomorphizing AI agents as 'coworkers' in the enterprise, as it degrades human oversight and accountability.
Why it matters
Anthropomorphizing agents dangerously lowers human vigilance, leading to unvetted errors in production workflows.
The take
This is a critical UX and system design insight. If you frame your agents as 'autonomous employees' on an org chart, users will over-trust them and fail to catch critical errors. Builders must design interfaces that preserve the user's sense of agency and oversight.
Do this
Design agent interfaces and system copy to emphasize tool-like utility and explicit verification steps rather than human-like colleague status.
Don't read this site daily. Get it in your inbox.
The daily brief and Sunday deep dive — distilled, scored, and opinionated. For builders only.