Opinion: Amazon’s AI talk is making Wall Street drool, but there seems to be a catch

It’s easy to get carried away thinking about the potential in generative AI, which can unlock more tech spending by companies looking to jump on the bandwagon. Yet corporate customers are still trying to find their footing with the nascent technology, as they play around with potential applications and get a better sense of their costs and benefits.

“The only certainty is that there will continue to be a high rate of change,” Amazon
AMZN,
+6.83%
Chief Executive Andy Jassy said on the company’s earnings call. “I think we are very early in generative AI….so many companies are doing all sorts of prototypes.”

Jassy, who led Amazon’s AWS cloud-computing business before he got the CEO nod, noted that many customers test models, like large language models for AI, and then plug those into their own software applications. “And what a lot of companies figure out quickly is that using…the large models and the large sizes ends up often being more expensive than what they anticipated.”

In other words, companies may be showing some interest in AI experimentation now, but they may not necessarily go through with all these projects over the long haul, or they could scale down the web services they purchase.

Plus, revenue for the cloud business can be hard to predict. “Deal volume tends to be lumpy…and it doesn’t perfectly distribute over a calendar year,” Jassy said.

That dynamic is already manifesting and could partly explain why Amazon came up about $100 million short of estimates with its cloud-computing revenue in the latest quarter. Jassy highlighted some new contracts that Amazon got for its AWS business in September but that won’t start until October, so they will be reported in the fourth quarter.

He also mentioned some “really big public-sector deals that won’t hit for a period of time.”

“All these deals don’t hit in a month,” he said. “They happen over a period of time.”

Additionally, as the economy has turned in the past year, deals are taking longer and companies are bringing more people into the decision-making process. “We are starting to see companies look forward more,” he added.

Then there was the dreaded talk of “optimization,” a term that refers to how companies are trying to get the most for their money when it comes to data-center assets. Jassy used the word a dozen times on the earnings call, joining executives at Alphabet Inc.
GOOG,
-0.03%

GOOGL,
-0.09%
and Microsoft Corp.
MSFT,
+0.59%
earlier in the week who also highlighted the trend.

Amazon’s biggest rival in cloud services, Microsoft, was the No. 1 public cloud services provider for 2022, with 16.2% market share, according to IDC. AWS, formerly No. 1 for years, was right behind it at No. 2, with a 13.5% share of the $545.8 billion worldwide cloud services market.

This year, it’s even feasible that Microsoft could take more share. Its Azure cloud business reported a better-than-expected quarter earlier this week, and the company has an investment in and partnership with OpenAI, the creator of ChatGPT. Microsoft also has promising early AI offerings, including Copilot for its GitHub development platform and for Office 365.

Microsoft’s latest results got a far better reception than those from Alphabet, which spooked analysts as the Google Cloud business slowed to its lowest growth rate since 2019.

Amazon shares were choppy in the extended session following Thursday’s report, but they ultimately ended up about 5% higher.

But investors cheering the company’s AI-fueled cloud opportunities should be aware that Amazon is not likely to experience hockey-puck-shaped growth.

Amazon may seem a big AI play itself, but it appears to be an inconsistent one, at least for now.