After Microsoft’s and Alphabet’s closely watched AI earnings, big questions remain

Satya Nadella, chief executive officer of Microsoft Corp., during the 2024 CES event in Las Vegas, Nevada, US, on Tuesday, Jan. 9, 2024.
Microsoft CEO Satya Nadella announced soaring cloud revenues off the company’s AI push—but also soaring costs.
David Paul Morris—Bloomberg via Getty Images

Yesterday saw a broad tech stock selloff, following Microsoft’s and Alphabet’s quarterly earnings on Tuesday.

Microsoft fell 2.7% and Alphabet 7.5%, while Apple (–1.9%), Amazon (–2.4%), and Nvidia (–2%) also got dragged down. As I wrote on Monday, this week’s results were always going to be seen as an indicator of the wider Big-Tech-AI sector’s prospects. So, is this as damning as it may appear at first?

Eh—frankly, the jury’s still out. Certainly, in the last quarter, both Microsoft and Google saw a substantial lift (20% and 26%, respectively) in cloud revenues, in both cases beating analyst expectations. Alphabet’s more substantial drop in its share price was largely down to disappointing ad revenues, but with both Alphabet and Microsoft, investors seem to have been a little put off by the massive infrastructural costs the companies forecast as the AI race continues this year. Both pared much of those stock losses this morning, as did their peers.

The fact is, we still don’t know exactly what we’re seeing here. Even though Microsoft hasn’t revealed sales or user figures for Copilot, its AI-powered chatbot, there does seem to be a good amount of uptake—but what are companies using it for? Everyone’s yelling at them, telling them to adopt AI or be left behind, so of course they’re experimenting. But will that translate to systematic usage, and to what degree?

In the words of Forrester analyst Lee Sustar, as quoted by the Financial Times, Alphabet’s and Microsoft’s customers are in “’buy AI now, figure out if it works later’ mode.” We can say for sure that lots of people are interested in AI and that building it out is hella expensive, but time will tell if we’re witnessing the effects of hype or something more sustainable.

Another thing worth watching is the PC market, in which vendors are tying revived uptake to the AI revolution. After all, a lot if not most of AI inference is going to take place on-device rather than in the data center. Qualcomm, which is preparing to take on both Apple silicon and Intel and AMD’s x86-based processors with its ARM-based Snapdragon X Elite chipsets, said yesterday that the promising product will appear in the middle of the year, before back-to-school season.

“We’re tracking to the launch of products with this chipset tied with the next version of Microsoft Windows that has a lot of the Windows AI capabilities,” Qualcomm CEO Cristiano Amon said in an earnings call. “We expect Snapdragon X Elite to set the industry benchmark for on-device gen AI and Copilot experiences in addition to leading performance and battery life for next-generation Windows PCs.” As The Register points out, hints of this AI-ified Windows version abound, but Microsoft hasn’t actually announced it yet.

How will Microsoft strike the local/cloud AI balance in Windows 12, or whatever it’s going to be called? And what will that mean for its revenues? Again, so many questions. More news below.

David Meyer

Want to send thoughts or suggestions to Data Sheet? Drop a line here.

NEWSWORTHY

Tesla environmental suit. Tesla’s been sued again, this time by over two dozen Californian counties, which say Elon Musk’s automaker improperly disposed of hazardous materials—paint materials, brake fluids, used batteries, etc.—on their turf. As Reuters reports, the civil penalties involved here could run to $70,000 per violation per day, which would add up to quite a lot, given that violations allegedly occurred at around 100 facilities, over a period of years.

Okta cuts. Another day, another layoff announcement. This time it’s access and ID management outfit Okta, which is letting go around 400 people, or 7% of its headcount. CEO Todd McKinnon, in a company email reported by TechCrunch: “To capture our massive potential and build an iconic company, we must be thoughtful about where we place our bets. This action is a proactive measure to help set the company up for long-term success.”

AI-powered robocalls. The Federal Communications Commission is set to make AI-generated robocalls illegal. FCC Chairwoman Jessica Rosenworcel: “AI-generated voice cloning and images are already sowing confusion by tricking consumers into thinking scams and frauds are legitimate. No matter what celebrity or politician you favor, or what your relationship is with your kin when they call for help, it is possible we could all be a target of these faked calls. That’s why the FCC is taking steps to recognize this emerging technology as illegal under existing law, giving our partners at state attorneys general offices across the country new tools they can use to crack down on these scams and protect consumers.”

ON OUR FEED

“I’m sorry for everything you have all been through … No one should go through the things that your families have suffered, and this is why we invest so much and we are going to continue doing industry-wide efforts to make sure no one has to go through the things your families have had to suffer.”

Meta CEO Mark Zuckerberg addresses families of teenage victims of online sexual exploitation, face-to-face at a Senate Judiciary Committee hearing. He made the surprise move after Sen. Josh Hawley (R-Mo.) asked if Zuckerberg would like to “apologize for what you’ve done to these good people.”

IN CASE YOU MISSED IT

If child safety policies ‘were working, we wouldn’t be here today,’ senators tell Zuckerberg, Yaccarino, and other CEOs in hours-long hearing, by Kylie Robison and Paolo Confino

ByteDance’s CEO is mad at his employees for missing the AI wave. He’s the latest Chinese tech executive to fret about being too slow to adapt, by Lionel Lim

The uncomfortable truth about AI’s impact on the workforce is playing out inside the big AI companies themselves, by Jason Del Rey

Tether still has nearly $5 billion in loans despite pledging to reduce exposure to zero, by Leo Schwartz

Adobe says its war on deepfakes could help combat the fake Taylor Swift porn problem, by Rachyl Jones

The fight against greenwashing starts with AI. Here’s why, by Anirban Bose (Commentary)

BEFORE YOU GO

Coinbase’s political hire. The U.S. crypto giant Coinbase has made an unusual appointment—its latest advisory council member is none other than George Osborne, the U.K.’s former chancellor of the exchequer (the British equivalent to the U.S. Treasury Secretary).

Well, perhaps not that unusual. It’s now more than five years since Facebook (as it was then) hired Nick Clegg as its chief lobbyist—Clegg was the U.K.’s Liberal Democrat deputy prime minister in the coalition government that governed the country in the first half of the 2010s; Osborne was one of his Conservative cabinet colleagues.

The former finance chief, who is these days also a banker among other things, yesterday trilled that he looked forward to working with Coinbase as it builds “a new future in financial services.” Coinbase chief policy officer Faryar Shirzad said he looked forward to “relying on [Osborne’s] insights and experience as we grow Coinbase around the world.” Which just goes to show: If you want a smooth-talking contacts machine to ease your international worries, look no further than the detritus of British politics.

This is the web version of Data Sheet, a daily newsletter on the business of tech. Sign up to get it delivered free to your inbox.