Despite growing dependence on artificial intelligence by professionals keen to stay ahead, a fresh academic study suggests that such usage may quietly backfire. Researchers at Duke University hihglights that employees using AI tools like ChatGPT or Gemini often face unfavorable judgments that erode their standing at work. The paper, published in the Proceedings of the National Academy of Sciences (PNAS), surveyed more than 4,400 individuals and found recurring signs of a social cost tied to workplace AI use.
Participants consistently viewed their AI-using peers as less competent, not as driven, and even somewhat lazy. These impressions weren't just imaginary fears; they shaped real-world interactions and reputations. As a result, some workers choose to hide their reliance on AI altogether, believing the act of disclosure might invite criticism or appear like a shortcut taken in place of effort.
While earlier research has focused on the functionality or ethics of these tools, this new study argues that the bigger blind spot lies in how coworkers perceive each other once AI enters the workflow. According to the authors, people don’t only worry about what AI does, they also worry how others react when they use it. This invisible pressure may be quietly blocking wider adoption.
The paradox is obvious. AI clearly helps with routine tasks, learning curves, and creative refinement. Yet, that boost in efficiency might come at a quiet cost. Many users now keep quiet, fearing they'd be seen as replacing hard-earned skill with shortcuts. Scholars called this tension a “social evaluation penalty,” which they believe is underexplored but increasingly influential.
Not long ago, stories of AI super users filled headlines: individuals who routinely use AI to polish content, manage workloads, or even just avoid dull assignments. They’re seen as forward-thinkers. But the same narrative, when observed inside a team or office, often triggers silent suspicion.
Elsewhere, researchers in Denmark published their own findings, noting that tools like Gemini or ChatGPT have yet to shift real-world wages or affect job hours meaningfully. Their conclusion ran counter to current assumptions, downplaying fears of a labor market shaken by generative AI. They argued that, at least so far, chatbot-driven disruption remains more theoretical than proven.
Still, the Duke study reveals something subtler. The most immediate barrier to AI isn’t its quality or access, it’s judgment from across the hallway. And in quiet corners of the workplace, that judgment speaks louder than algorithms.
Image: DIW-Aigen
Read next:
• Research Reveals How AI Chooses Words by Memory, Not Rules
• Generative AI Platforms See Remarkable Engagement, With Users Spending 6+ Minutes per Session
Participants consistently viewed their AI-using peers as less competent, not as driven, and even somewhat lazy. These impressions weren't just imaginary fears; they shaped real-world interactions and reputations. As a result, some workers choose to hide their reliance on AI altogether, believing the act of disclosure might invite criticism or appear like a shortcut taken in place of effort.
While earlier research has focused on the functionality or ethics of these tools, this new study argues that the bigger blind spot lies in how coworkers perceive each other once AI enters the workflow. According to the authors, people don’t only worry about what AI does, they also worry how others react when they use it. This invisible pressure may be quietly blocking wider adoption.
The paradox is obvious. AI clearly helps with routine tasks, learning curves, and creative refinement. Yet, that boost in efficiency might come at a quiet cost. Many users now keep quiet, fearing they'd be seen as replacing hard-earned skill with shortcuts. Scholars called this tension a “social evaluation penalty,” which they believe is underexplored but increasingly influential.
Not long ago, stories of AI super users filled headlines: individuals who routinely use AI to polish content, manage workloads, or even just avoid dull assignments. They’re seen as forward-thinkers. But the same narrative, when observed inside a team or office, often triggers silent suspicion.
Elsewhere, researchers in Denmark published their own findings, noting that tools like Gemini or ChatGPT have yet to shift real-world wages or affect job hours meaningfully. Their conclusion ran counter to current assumptions, downplaying fears of a labor market shaken by generative AI. They argued that, at least so far, chatbot-driven disruption remains more theoretical than proven.
Still, the Duke study reveals something subtler. The most immediate barrier to AI isn’t its quality or access, it’s judgment from across the hallway. And in quiet corners of the workplace, that judgment speaks louder than algorithms.
Read next:
• Research Reveals How AI Chooses Words by Memory, Not Rules
• Generative AI Platforms See Remarkable Engagement, With Users Spending 6+ Minutes per Session
