Meta’s embrace of AI is making its employees miserable

SAN FRANCISCO – In an internal post in April, Meta told its US employees that it was making a change that would affect tens of thousands of them.

What employees typed on their computer, how they moved their mouse, where they clicked and what they saw on their screen would be tracked, Meta said. The goal, the company said, was to capture employee data so Meta’s artificial intelligence models could learn “how people actually complete everyday tasks using computers”.

Many workers immediately revolted. In online comments, they blasted the tracking as a privacy violation, calling it anti-social and callous.

“This makes me super uncomfortable,” an engineering manager wrote in response to the announcement. “How do we opt out?”

“There is no option to opt out on your corporate laptop,” replied Mr Andrew Bosworth, Meta’s chief technology officer. Employees reacted by posting more than 100 angry and surprised emojis, according to the messages.

Mr Mark Zuckerberg, Meta’s chief executive, has staked the future of his company on AI by weaving the powerful technology into apps like Facebook and Instagram and spending hundreds of billions of dollars on developing AI models and data centres.

But as the Silicon Valley company tries to transition from an internet company to an AI organisation, its embrace of the technology has been awkward and, at times, downright ugly.

Meta is pushing its 78,000 employees to adopt AI tools and factoring their use of the technology into performance reviews. The company is also tracking employees’ computer work to feed and train its AI models. At the same time, it is cutting jobs to offset its AI spending, saying in April that it would slash 10 per cent of its workforce.

That has led to anger and anxiety as employees await news of whether they are affected by the layoffs, which are scheduled for May 20, according to 11 current and former Meta employees.

Some said they no longer saw Meta as a place for a long career. Others were looking for new jobs or trying to signal that they wanted to be laid off so they could receive severance pay, the current and former employees said.

The angst at Meta offers a preview of what may happen at other tech companies as they increasingly incorporate AI into their workplaces.

Microsoft, Block and Coinbase have recently announced layoffs or buyouts as AI has reshaped work. The technology has been particularly disruptive at tech companies because AI tools are good at generating code, which was typically written by the software engineers who underpin many digital businesses.

“AI can potentially make everyone a better coder and help them do way more things with fewer resources, but as a result, it also brings more intensity to the daily life of the worker,” said assistant professor of information systems Leo Boussioux at the University of Washington.

“There is no playbook for AI in the workplace yet,” he added.

In a statement, Mr Tracy Clayton, a Meta spokesman, said the purpose of the new employee tracking programme was to train the company’s AI products. “There are safeguards in place to protect sensitive content, and the data is not used for any other purpose,” he said.

Meta began reorienting itself around AI not long after OpenAI released the ChatGPT chatbot in 2022. Last summer, Mr Zuckerberg spent billions to create a lab for “superintelligence” – a futuristic form of AI that can act as the ultimate personal assistant – and overhauled its AI division.

In March, Meta organised “AI Transformation Weeks” for its workers, five current and former employees said. The aim was to teach employees how to use AI coding tools and AI agents, which are digital assistants that can do tasks by themselves. Product designers, who style the way products look and function, were told to use AI to try coding, and software coders were told to use AI to try designing products.

Meta also introduced internal dashboards to track employees’ consumption of “tokens”, a unit of AI use that is roughly equivalent to four characters of text, four people said.

Some said the dashboards were a pressure tactic to encourage competition with colleagues. That led some employees to create so many AI agents that others built additional agents to locate and evaluate them, two people said. 

On April 17, Reuters reported that Meta would soon lay off 10 per cent of employees. Workers fretted over whether they had been training their AI replacements, three people said.

Some employees have since shared layoff guides and nihilistic memes. “It do not matter,” read one meme shared internally.

Employees have created at least three websites counting down to the May 20 layoffs, with one website’s header reading, “Big Beautiful Layoff”, a play on the 2025 domestic policy law that US President Donald Trump called the “One Big Beautiful Bill”.

Meta has hinted at more changes.

“We don’t really know what the optimal size of the company will be in the future,” Ms Susan Li, the chief financial officer, said during a call with investors in April. “There’s a lot of change right now, with AI capabilities advancing rapidly.”

The day after the investor call, Mr Zuckerberg held a company-wide question-and-answer meeting, along with other executives.

During the meeting, a recording of which the Times reviewed, Mr Zuckerberg said Meta was not gathering data on employees for “surveillance or performance tracking or anything like that”.

Instead, the data would train AI on “how smart people use computers to accomplish tasks”, he said.

“I think we know that AI is one of the most competitive fields, probably in history,” Mr Zuckerberg added. NYTIMES