Meta has played a card that could be highly uncomfortable for employees as part of its push to expand its share of the AI market. It said it would track what employees view and do on their computers. The stated rationale is to secure more data for AI models.

It said it would track everything employees do on their computers, including typing and mouse movements, as well as what they click and view on screen, so Meta AI models can learn how people actually complete everyday tasks on computers.

Meta employees have, as expected, pushed back.

According to a recent New York Times report, Meta employees vented their complaints online. Even so, Meta CTO Andrew Bosworth (앤드류 보스워스) made it clear there would be no exceptions.

The measure comes as Meta continues bold investment after betting its future on AI. Meta is pouring astronomical sums into data centres and AI model development, but it is still not treated on a par in the AI race with OpenAI, Anthropic and Google. Even so, Meta's push into AI is increasingly in attack mode. It is also accelerating efforts to reshape the company around AI beyond bold investment.

As part of that, it has strongly urged the roughly 78,000 employees to use AI and decided to reflect that in performance evaluations. It also recently announced it would cut 10 percent of its workforce to offset AI spending.

The controversial measure to track how employees use computers is an extension of that approach. For Meta employees who had been anxious about whether they might be targeted for layoffs, anger at the company's announcement was a natural result. Many Meta employees responded that tracking was anti-social, inhumane and a violation of privacy, and it was reported that not a few said morale was dropping and they were considering changing jobs.

What is unfolding at Meta is another scene showing rising anxiety among salaried workers as tech companies, large and small, announce layoffs one after another citing AI. AI-driven layoffs have become a particularly sensitive issue at tech companies because AI still delivers the best results in coding, a core tech-company specialty. As various stories spread among developers about doing more with fewer people thanks to AI, layoffs have become a tool that frequently accompanies the goal of boosting AI productivity.

ㆍ[Tech Inside] Why programmers embrace AI even as it disrupts coding? ㆍLarge-scale layoffs at tech firms under the banner of AI... concerns over a "risky gamble" as well

In Meta's case, it has gone beyond trimming headcount because of AI and appears to be aiming to turn the organisation itself from an internet company into an AI company, moving to rapidly introduce AI across the organisation.

With no proven standard for how work should be done in the AI era or what is good, Meta is putting forward bold experiments one after another first. At this point, Meta's AX (AI transformation) process is not neat. Borrowing the New York Times' wording, Meta's move to adopt AI across the organisation is chaotic and at times even awkward to watch.

Even so, the scenes playing out at Meta around AI seem likely to become one example that shows in advance what changes AI may bring to what people do at companies and how businesses are run.

ㆍLayoffs or keeping jobs while expanding results... corporate HR at a "crossroads" in the AI era

Meta held an event called AI Transformation Week for employees in March. It was aimed at showing them how to use AI coding tools and AI agents. At the event, product designers had to try coding with AI, and employees who had been coding software had to try product design using AI.

Meta also introduced an internal dashboard that tracks how many tokens employees use. In the AI world, tokens are seen as a key indicator of how much AI is being used. Among some Meta employees, the token-tracking dashboard was said to be viewed as a pressure tactic to stoke internal competition. The New York Times reported, citing 2 sources, that as some employees created too many AI agents, other employees even released agents that find agents and agents that evaluate agents.

With large-scale restructuring compounded by an announcement that it will track computer use, the mood among Meta employees appears unsettled. It is unclear whether that unsettled mood will subside easily. Judging from what Meta executives are saying, there may be even more change ahead. Meta executives' position is that in a climate of intensifying competition around AI, bold moves are unavoidable regardless of what employees think.

Meta CFO Susan Li (수전 리) said on a recent quarterly earnings conference call, "I don't know what the optimal company size is," adding, "As AI capabilities are advancing quickly, I think a lot of change is happening right now."

According to a transcript obtained by the New York Times, Meta CEO Mark Zuckerberg (마크 저커버그) said at an all-hands meeting the day after the conference call, "We do not collect employee data for surveillance or performance tracking purposes, or for similar purposes. The data collection will be used to train AI on how smart people use computers to do their work." He added, "We all know that AI is one of the most competitive fields in history."