Senior figures across the technology industry have repeatedly argued that artificial intelligence will reduce workloads, create more leisure time, and fundamentally reshape how much effort humans need to put into their jobs. Yet employees working inside some of those very same companies say the reality looks nothing like that promise. Reports from staff at major tech firms describe punishing schedules of up to 90 hours per week, raising pointed questions about whether the industry is living by its own gospel.
The Promise: AI as a Path to Easier, Shorter Working Lives
For several years, leading voices in the technology sector have championed artificial intelligence as a tool that will liberate workers from repetitive, time-consuming tasks. The argument goes that as AI handles more of the routine workload — drafting documents, analysing data, writing code — employees will be freed to focus on more creative, strategic, and personally fulfilling work, ideally in fewer hours. These claims have been made not just in conference keynotes but in investor briefings and public interviews, helping to frame AI adoption as a broadly positive social development rather than purely a cost-cutting measure.
The vision has obvious appeal. Decades of productivity research have shown that overwork tends to reduce output quality and harm employee wellbeing, so the idea that a technological leap could simultaneously make businesses more efficient and workers less stressed carries genuine weight. Policymakers and economists have also engaged with the idea seriously, with some analysts suggesting AI could eventually contribute to a shorter standard working week across whole economies. That broader context makes the credibility of these claims all the more important to examine.
The Reality: Long Hours Inside the AI Industry Itself
What makes the current situation particularly striking is not just that tech workers are stretched thin — demanding schedules have long been a feature of the industry — but that the companies making the loudest claims about AI-driven liberation appear to be among those where staff are working the longest hours. Employees at some major tech firms have described schedules stretching to 90 hours a week, a level of intensity that is difficult to reconcile with the promise of technology freeing up human time. If AI is genuinely transforming productivity inside these organisations, workers are not yet seeing that translate into more time away from their desks.
There are several possible explanations for this disconnect. One is that AI is generating new categories of work at least as fast as it eliminates old ones — tasks like prompt engineering, model oversight, data labelling, and rapid iteration on AI-assisted outputs can expand to fill available time. Another is that competitive pressure within the tech sector is intensifying as companies race to deploy AI products, meaning any efficiency gains are immediately absorbed by accelerated ambition rather than reduced working hours. In this environment, AI becomes a tool for doing more, not for working less.
A Credibility Problem With Wider Consequences
The contradiction between what tech leaders say and what their own staff experience is not merely a matter of internal company culture — it has implications for how AI is perceived and regulated more broadly. When industries advocate publicly for a technology while simultaneously failing to model its supposed benefits in their own workplaces, it can erode trust among the general workforce and policymakers who are being asked to support or accommodate that technology's spread. Workers in other sectors observing this pattern may reasonably question whether AI adoption in their own industries will genuinely improve their conditions or simply intensify existing pressures.
There is also a governance dimension worth considering. Labour regulators and trade unions in various countries are beginning to pay closer attention to how AI affects working conditions, not just employment levels. If evidence continues to mount that AI tools are being used primarily to extract more output from the same or fewer workers — rather than to improve their quality of life — calls for regulatory intervention around working hours, workload monitoring, and employee rights in AI-heavy environments are likely to grow louder. The tech industry's ability to shape that conversation will depend heavily on whether its own house appears to be in order.
Why it matters
The gap between AI's promised benefits for workers and the lived experience of employees inside the tech industry matters because it shapes public trust in a technology that is rapidly being adopted across nearly every sector of the economy. If the companies building and selling AI cannot demonstrate that it improves working conditions for their own staff, the broader argument that AI is good for workers becomes much harder to sustain. This tension is likely to influence policy debates around labour rights, productivity expectations, and workplace regulation for years to come.
Common questions
Why are tech workers putting in such long hours if AI is supposed to save time?
Efficiency gains from AI tools appear to be absorbed by increased ambition and faster product cycles rather than reduced working hours. Companies racing to stay competitive in the AI space are using productivity improvements to do more, not to give employees a lighter schedule.
Does this mean AI won't actually reduce working hours for most people?
It is too early to say definitively, but the current evidence from inside the tech industry suggests that without deliberate policy choices or cultural change, AI productivity gains tend to intensify work rather than shorten it. Whether that pattern holds across all industries will depend significantly on how employers, regulators, and workers negotiate the terms of AI adoption.
