Tech leaders say AI means less work - their staff say they work up to 90 hours a week
Despite promises from major tech executives that artificial intelligence will reduce human labor, employees inside leading AI firms report grueling work hours and expanding workloads.
Stock photo for illustration only, not from the actual event
- Tech executives have long claimed that AI will eventually reduce working hours and introduce a four-day work week.
- A former OpenAI employee revealed the company never actually trialled the four-day work week it recommended to others.
- Staff at OpenAI and Anthropic report logging up to 90 hours a week during intensive project sprints.
- Research from UC Berkeley indicates AI usage causes employees to work faster and handle a broader scope of tasks.
For years, executives at companies pouring hundreds of billions of dollars into artificial intelligence development have insisted the technology will ultimately cut down the time people spend working. Four years ago, an engineering director at Google predicted that AI would deliver a four-day work week by 2025.
Earlier this year, just one year after that prediction, OpenAI urged businesses to test out a four-day work week with no reduction in pay, claiming AI would soon speed up human labor significantly. However, a former OpenAI technical employee who left last year told the BBC that the firm never actually trialled the four-day work week it suggested others should try.
Stock photo for illustration only, not from the actual event
Other companies have also pushed the narrative that AI will reduce necessary working hours. While Meta has laid off one in ten employees, Anthropic CEO Dario Amodei warned that increased productivity could lead to even broader job losses. Despite these claims, workers inside these exact tech companies report clocking far more hours than the typical five-day, 40-hour work week.
US tech workers are typically well-compensated, and while long hours are common in fields like investment banking and medicine, a 14-hour workday was not historically the norm in technology. Nevertheless, the former OpenAI employee stated they regularly put in at least 70 hours a week, far exceeding their previous tech roles. They now work at an AI-focused start-up, working closer to 50 to 60 hours a week outside of sprints.
This dynamic reflects the automation paradox, where efficiency gains from new tools often lead to higher demands, increased verification requirements, and expanded responsibilities for workers rather than actual leisure time.
A sprint refers to a common period in tech companies where staff work extended hours ahead of a product release. Inside companies pushing hard on AI projects, these sprints can become extreme.
"Such sprints can stretch on for many weeks and top 90 hours of work in a seven-day period."
Tech workers interviewed by the BBC
At OpenAI and Anthropic, workers told the BBC that sprints can stretch on for many weeks and top 90 hours in a single week. Neither company responded to BBC requests for comment. Meanwhile, Meta workers reported being abruptly drafted onto urgent AI teams without a choice, working through the night, weekends, and remaining constantly on call.
Stock photo for illustration only, not from the actual event
Meta's current AI projects involve building software engineering tools and model infrastructure to measure AI task replication, work that employees described as endless. Furthermore, research from UC Berkeley tracking workers at a US tech company found that employees using AI worked at a faster pace, took on broader tasks, and extended work into more hours of the day.
Source: BBC Business
Found something wrong in this article? Report an issue with this article
Comments
Leave a Comment