Remember how, during and immediately after the pandemic lockdown, many companies promised to usher in a new era of work-life balance and flexible working? According to the latest research from Microsoft, the opposite is now true, with most people working more than 12 hours and even extending into the weekend. This is impacting productivity, and while AI can improve the situation, it may also make it worse.
Microsoft's June 2025 Work Trends Index Special Report warns that more and more people are now stuck in seemingly endless workdays. The working day starts at 6 a.m. and lasts until after 8 p.m., including Saturday and Sunday.

The findings, based on trillions of aggregated and anonymized Microsoft 365 productivity signals around the world, show that 40% of users who go online at 6 a.m. are checking email to determine their priorities for the day.
Half of all meetings are held between 9-11 a.m. and 1-3 p.m., the most productive times of the day, squandering people’s natural midday productivity peak. 11 a.m. is also a peak time for messaging activity, as real-time messages, scheduled meetings, and constant application switching all converge here.
For many, work continues late into the night. Microsoft found that meetings held after 8 p.m. increased 16% over last year. Additionally, employees now send an average of more than 50 messages outside of core work hours, and nearly a third (29%) of active employees check their inbox by 10 p.m.
There was little respite over the weekend. About 20% of employees check email before noon on Saturday and Sunday, and more than 5% process email on Sunday evening.

Data shows that the average employee receives 117 emails and 153 Teams messages per day. This means employees using Microsoft 365 are interrupted by meetings, emails, or notifications every 2 minutes.
Not surprisingly, almost half of employees and more than half of leaders feel their jobs are disorganized and scattered.
Microsoft says that artificial intelligence can not only provide a way out of this endless workday, but also speed up the operation of existing systems. The company recommends deploying AI and agents to streamline low-value tasks and focus on the "80/20" principle, where 20% of the work delivers 80% of the results. Additionally, the company recommends moving from rigid organizational structures to agile, results-oriented teams, supplemented by artificial intelligence.
Artificial intelligence agents are repeatedly highlighted in the report as the solution to these endless workdays. Of course, not mentioning that they might put humans out of work.
This isn't the first time research has shown that working long hours, especially without stopping, can have a negative impact on productivity. Another report found that the most productive workers had a 75/33 work-break ratio: 75 minutes on the job and 33 minutes off. The report also noted that working in an office, where people stop to do things like talk to colleagues or even move around, is more productive than working non-stop at home.