Microsoft posts $36.6bn quarterly profit days after offering buyouts to 60,000 staff: payroll is being moved into AI compute

Microsoft says its profit for the three months to the end of September rose to more than $36.6bn (£27.5bn). According to the BBC, the results came just days after the company said it would offer buyouts to around 60,000 employees. That is more than a quarter of a workforce of over 220,000. Revenue reached $92.3bn, up 19% on a year earlier. Microsoft Cloud, its range of enterprise technology services, grew revenue by 30%. The shares rose almost 2.9% in after-hours trading in New York.

The official reason for the job cuts matters here. The BBC reports that Microsoft is cutting costs as it pours billions of dollars into artificial intelligence. The company describes the scheme as a voluntary redundancy offer for eligible employees in Canada, India, the UK and the US. It says the aim is to adapt its workforce to "changing needs" and to create "greater flexibility", with financial support and career planning for those who leave. One detail is often lost in the headlines. This is an offer, not a list of confirmed dismissals. How many of the 60,000 actually leave remains to be seen.

The wider context explains the timing. Amazon, Meta and Alphabet are expected to spend more than $700bn this year on AI chips and infrastructure such as data centres. Investors are watching every set of results to judge whether that spending is sustainable. On the same day, they gave very different answers. Alphabet reported a 25% drop in quarterly net profit, said it would keep spending heavily on AI, and saw its shares fall 9.75%. Microsoft, by contrast, could point to cloud revenue up 30%. Even so, the company itself listed its "substantial investments" in AI among the risks to its finances.

Our reading: this is not a company in trouble. It is a company making a choice. A firm earning more than $36bn in a quarter does not need buyouts for more than a quarter of its staff in order to survive. It is moving costs from payroll to computing infrastructure, and the market has rewarded the trade. This fits a pattern we have described before. For the big AI players, the business increasingly behaves like capital-intensive infrastructure, and labour is treated as one more cost to optimise against the bill for chips and power. The contrast with Alphabet adds a useful nuance. Markets are not rewarding AI spending blindly. They reward it when it already shows up as revenue, as it does in Microsoft's cloud business. They punish it when it eats into profit without a clear return.

In the short term, we won't pretend this is anything other than hard. Tens of thousands of people are being invited to leave a company that has never been more profitable. That is the uncomfortable side of the transition we have warned about: gains go to capital, and adjustment costs fall on workers. Satya Nadella says AI is delivering "real-world impact across every industry". If that is true, the test is whether some of that impact reaches the people who leave, through retraining, real career support and new roles, and not just the share price. In the long run, we remain optimistic. Productivity that frees people from routine work can lead to abundance and work people actually choose. But that outcome isn't automatic. It depends on policy, on corporate decisions and on how the gains are shared. Microsoft has just shown it can afford to do this well. Now it has to show that it will.