Have you caught yourself feeling unusually confident at work — or unsure about errors slipping through your workflow?
You may not be imagining it.
A new report from the Work AI Institute, produced with researchers from universities including Notre Dame, Harvard, and UC Santa Barbara, and released on Wednesday, said that AI is turning ordinary office workers into people who feel smarter and more productive while their underlying skills slowly erode.
“AI is putting expertise into our hands in a way that’s not always predictable,” Rebecca Hinds, head of the Work AI Institute at workplace search company Glean and the report’s coauthor, told Business Insider.
“There’s often this illusion that you have more expertise, more skills than you actually do,” she said. “Even if you’re very well aware you’re using the technology, it’s often unclear where your knowledge ends and where the technology begins.”
Hinds drew a parallel to the rise of search engines, when people began to mistake easy access to information for genuine understanding.
With generative AI, she said, that illusion is even more powerful — and the risks are higher.
When AI erodes skills instead of sharpening them
Hinds said these risks are most obvious in creative and knowledge-intensive roles.
Workers are increasingly using AI to beat the “blank page,” she said, and generate first drafts of writing.
That speeds things up, she said, but it also strips away the messy, time-consuming work of wrestling with ideas.
“The more you poke holes in it, the more it feels yours and the more you commit to it, and the more you’re able to fight for it in a meeting if someone pushes back,” she said.
“That process is highly inefficient,” she added, “but it’s also really healthy.” And if workers lean too heavily on AI to skip it, “your skills are going to atrophy.”
The report suggested that AI can create either a “cognitive dividend” or “cognitive debt.”
Used intentionally, as a partner in domains where you already have expertise, it can free up time and sharpen judgment. However, used as a reflexive shortcut, it leads to weaker skills and misplaced confidence, it said.
Early-career workers are most exposed
Hinds said that the roles with the highest exposure are early-career jobs.
Those are the roles that traditionally function as apprenticeships: junior developers learning from senior engineers, entry-level marketers learning how to build campaigns, and young analysts learning how to structure a model from scratch, she said.
If those tasks are automated away or if juniors rely entirely on AI to do them, they may never develop the underlying skills they need to advance, she said.
Hinds said leaders are often unintentionally exacerbating the illusion-of-expertise problem.
A big red flag, she said, is “organizations stack-ranking employees based on how many times they’re clicking an AI tool as a marker of AI adoption or AI productivity or AI success.”
In some companies, usage metrics are tied directly to performance reviews.
Employees “are incentivized to click the tool more rather than invest in a deep understanding of the tool,” she added.
Instead, she said, companies should tie AI to existing business goals — quality, customer satisfaction, innovation — and measure whether it actually improves those, not just how often it’s used.
How to avoid becoming an ‘AI-powered amateur’
Hinds doesn’t think the solution is to shun AI. She thinks it’s to be far more deliberate about its use.
She recommended three questions for workers and leaders:
- What roles should stay deeply human? Identify the parts of your job that build judgment, creativity, and motivation — and resist fully automating those.
- Where is AI truly adjacent? Use AI first in areas close to your existing expertise, not as a shortcut into domains you don’t understand.
- What are you measuring? Focus less on how often people use AI and more on whether it’s improving real outcomes.
AI “does not magically transform you as a leader,” Hinds said. “More often, it amplifies what already exists within the organization.”
Read the original article on Business Insider
The post AI is giving workers the illusion of expertise — and quietly making them worse at their jobs, researchers say appeared first on Business Insider.
