AI’s potentially disruptive impact on computing platforms

Unlock Editor's Digest for free

FT editor Roula Khalaf has chosen her favorite stories in this weekly newsletter.

General euphoria over generative artificial intelligence has gripped Wall Street. But the technology had little to do with the strong results reported by most of America's largest technology companies in recent days.

Understanding where technology is starting to produce real business outcomes and where it is not will be key to differentiating AI winners from AI losers in the coming months and years.

For example, consider the business recovery that the largest cloud computing platforms have experienced this year. Last week, growth at Microsoft and Google's cloud computing divisions accelerated again, raising hopes that AI is starting to have a noticeable impact. This week, the upbeat mood was further boosted by the arrival of cloud market leader Amazon Web's services.

However, recent results question the extent to which this rebound in growth reflects increased spending on generative AI, how sustainable such spending turns out to be, and whether new AI services provides little insight into how much it costs to provide.

Cloud growth has declined significantly over the past two years. Many customers who experienced skyrocketing cloud prices during the pandemic have put the brakes on new spending as they look for ways to be more cost-effective.

This hiatus, which tech companies euphemistically refer to as a period of “optimization,” squandered one of the industry's biggest drivers of expansion. AWS revenue growth has fallen from 40% at the end of 2021 to a relative nadir of 12% after 18 months.

A rebound to 17% in recent quarters indicates that the indigestion caused by the early cloud spending binge is largely a thing of the past. According to Andy Jassy, ​​Amazon's chief executive and former head of cloud, this is a return to an earlier status quo, where the move to the cloud was driven by a desire to reduce IT costs. Only 15% of his company's IT workloads are on the cloud, and he argues that it will take a long time for this trend to continue.

AI is not the main force here, but it is definitely becoming a factor, if only a little. Most obviously, Microsoft's annual revenue from generated AI is now generally said to be around $4 billion, but Jassy also said that AI has become a “multi-billion dollar” business for AWS. .

It's unclear how fast these AI revenues will grow or how large the market will be. Customers are rushing to train new AI models and try out the new services they enable. But until this period of mass experimentation has passed, it is difficult to predict how much value a new technology will create or how much customers are willing to pay for it.

While the timing of payback is uncertain, the costs are very real. Alphabet, Amazon and Microsoft are expected to spend a combined $150 billion in capital this year, more than $40 billion more than they spent last year. These are huge down payments on the promise of a coming technology boom.

Reducing depreciation costs by extending the expected useful life of all this new data center equipment removed some of the large investment buildup for all three companies. Alphabet, for example, boosted its operating profit by nearly $4 billion last year by extending the life expectancy of its servers and network equipment to six years and spreading the cost of purchasing new equipment over time.

Another factor offsetting some of the pain is that cloud companies claim they can tie their investments more closely to expected short-term returns from customers lining up to try new technology. is. This helps explain why Wall Street has been so eager to embrace the recent uptick in investment from cloud companies.

What is even more unclear is whether the wave of generative AI will be disruptive enough to upset the balance of power in the cloud industry, which has seemed remarkably stable in recent years. AWS's revenue is $100 billion a year, probably double that of Microsoft's Azure cloud platform. Google is even further behind.

It's no surprise that customers are conservative in moving critical data and IT workloads between clouds, and AWS is rushing to build out AI capabilities. But Microsoft's early lead thanks to its partnership with OpenAI is a big factor in Azure's current growth rate of 31%, nearly double AWS's 17%.

This will be a long race, with every chance to reset the competitive dynamics between the tech giants.

richard.waters@ft.com