The first signs of fatigue come from the people who embrace AI the most

The most enticing narrative in American work culture right now is not that AI will take your job. It’s that AI will save you from it.

This is the version that the industry has spent the last three years selling to millions of nervous people eager to buy. Yes, some office jobs will disappear. But for most other roles, the argument goes, AI is a force multiplier. You will become a more capable and indispensable lawyer, consultant, writer, programmer, financial analyst – etc. Tools work for you, work less, everyone wins.

But a new study published in the Harvard Business Review followed this hypothesis to its actual conclusion, and what it found is not a productivity revolution. He concludes that companies are at risk of becoming burnout machines.

As part of what they described as “ongoing research,” UC Berkeley researchers spent eight months inside a 200-person technology company observing what happened when workers truly embraced AI. What they found across more than 40 “in-depth” interviews was that no one was stressed at this company. No one was asked to hit new targets. People just started doing more because the tools made them feel doable. But since they were able to do these things, work began to bleed through during lunch breaks and late in the evening. Employees’ to-do lists expand to fill every hour the AI ​​frees them up, and then they keep working.

As one engineer told them: “You thought that maybe, because you can become more productive with AI, then you can save some time, and you can work less. But in fact, you are not working less. You are working the same amount or even more.”

On the Hacker News tech industry forum, a commenter had the same reaction, writing: “I feel like this. Since my team jumped into the AI-in-everything way of working, expectations have tripled, stress has tripled and actual productivity has only gone up 10%. It’s like leadership is putting enormous pressure on everyone to prove that their investment in AI is worth it, and we’re all feeling the pressure to try to show that when we’re actually having to work longer hours to do so.”

It’s fascinating and also disturbing. The debate over AI and business has always gotten stuck around the same question: Are the gains real? But very few people stop to wonder what happens when they are.

The researchers’ new findings are not entirely new. A separate experiment last summer found that experienced developers using AI tools took 19% longer on tasks while believing they were 20% faster. Around the same time, a National Bureau of Economic Research study tracking the adoption of AI in thousands of workplaces found productivity gains of just 3% in time saved, with no significant impact on earnings or hours worked in any occupation. Both studies were selected.

This may be difficult to dismiss because it does not challenge the premise that AI can augment what employees can do on their own. He confirms this, and then shows where all this reinforcement actually leads, which is “exhaustion, exhaustion, and an increasing sense that work is difficult to step away from, especially as organizational expectations for speed and responsiveness rise,” according to the researchers.

The industry is betting that helping people do more will be the answer to everything. It may turn out to be the beginning of a completely different problem.