Last week, Meta revealed it is using a brand new source of vital data to train its in-house AI agents: by tapping the actions of thousands of workers on its own staff. It sounds smart—AI training data is hard to get, and expensive. But alarm bells immediately rang because the data was coming from tracking what workers were doing on their computers, down to capturing screenshots and tiny mouse movements. If you’re at all wary of trusting employers you’ll be unsettled by this news. Or, using some younger generation slang, the idea will give you “the ick.”
A spokesperson tried to explain the rationale to Reutersnoting that if the company is going to build AI “agents to help people complete everyday tasks using computers,” it makes sense that its models “need real examples of how people actually use them—things like mouse movements, clicking buttons, and navigating dropdown menus.” Meta, in an internal memo, promised that the data wasn’t going to be used for any kind of performance reviewand was only going to be fed into the company’s data-hungry AI tools.
But immediately Meta employees were suspicious of the idea and complained about it loudly, reports said. In a discussion about the news on Reddit, users’ comments were almost entirely negative. “This is cosmic horror levels of dystopia,” one wrotecommenting on the issue of training AI. Another predicted that “more tracking software and technology will come to all fields of work.” Others referenced Meta’s data-scraping thirst, even for its own users’ information, and one user suggested to the workers that “If they are smart, they will sabotage the data.” More damningly, another concluded “This is AI. It’s your replacement and you are to train it.”
In an email to Inc., Eva Chan, a career expert at Resume Genius underlined that this mistrust of employers’ motivation relating to AI runs deep, particularly for younger workers—the digitally native Gen-Z. In data from a survey shared by the resume-helper company, 49 percent of Gen-Z staff responded that they were already concerned that AI may lead employers to make unfair or biased decisions affecting them at work. Worse, nearly a quarter (23 percent) said that using AI at work has harmed their mental health and over one in three (37 percent) said that AI makes them feel replaceable.
“When you collect granular behavioral data from employees, the line between ‘training data’ and ‘performance signal’ is paper-thin, regardless of stated intent,” Chan noted. With so much anxiety about the role of AI tools, and questionable ethics relating to employee surveillance software, companies deploying tools like this “need to be far more transparent about where the data starts and stops—or they’ll lose the trust of the workforce they’re trying to build around,” Chan explained.
When this happens with Gen-Z staff, the distrust situation could easily be amplified, triggering skittish Gen-Z workers to quit and search for better work elsewhere. The impact of high Gen-Z staff turnover may also, ironically, impact the ability of a company to smoothly roll out advanced automation or AI tech. Numerous reports have highlighted that Gen-Z’s digital smarts make these young workers perfect ambassadors, and indeed trainers, for older workers who are more wary of this brand new tech.
As a report in Fortune recently warnedsenior, often older, managers face some serious risks in unsettling Gen-Z workers—with the generations’ famous “pout” and “stare” trends as warnings about mismatched expectations of traditional-thinking managers and young AI-savvy staff. Gen-Z now comprises about 30 percent of the workforce, but Baby Boomers, “who currently occupy a significant share of managerial and executive roles” think differently, Fortune noted. They “built their careers inside a workplace culture defined by formality, hierarchy, and personal relationships cultivated face-to-face.” Coming face-to-face with Gen-Z’s different ideas about workplace etiquette challenges older workers’ expectations about hierarchy. Recent reports also note that Gen-Z is already dubious about trusting older colleagues.
