Tesla, Meta, IBM: 3 AI Signals From Earnings Calls
Earnings calls at Tesla, Meta and IBM highlighted three different AI issues: Tesla cut its capital spending, Muse users shop with an average value above $200, and big companies are starting to question the returns on AI investment. Executives explained the challenges and opportunities they see.
Tesla’s capital spending this year fell to $20 billion from a planned $25 billion, while Meta reported that Muse users shop with an average value above $200. The earnings calls of the three companies, Tesla, Meta and IBM, showed different challenges in applying AI, from training robots to credit risk and measuring returns on investment.
Yahoo Finance reported three moments from the Tesla, Meta and IBM earnings calls held on Wednesday evening. Attention centered on how companies are developing AI while controlling the costs and effects of its use.
The differences are real. Tesla faces the need to train its Optimus robot, Meta is seeing shopping activity through its Muse AI agent, and IBM raised large companies’ concerns about returns on AI spending.
Tesla trims capital spending plans
Tesla lowered its capital spending forecast for the year to $20 billion, from the $25 billion it gave three months earlier. One factor that appears to have played a role is the need to give Optimus more training before the robot is released into the real world.
Tesla CEO Elon Musk asked investors to picture Optimus as a baby that is still learning. He said Tesla does not have a large fleet of robots to collect training data the way it does for cars, so the robots need to train through their own activities.
“With Optimus, we have to bootstrap it by making the robots and having them do self-play,” Musk said. He compared the learning process to a small child who at first cannot do much, then develops through play.
According to Musk, Optimus needs to play at scale to be able to perform tasks well in the real world. The explanation gives context to the spending cut: the need to train robots is part of the company’s effort to develop the technology, although the Yahoo Finance report does not detail how Tesla’s budget is divided.
Meta sees shopping activity on Muse
At Meta, CEO Mark Zuckerberg said shopping is the most common use of the Muse AI agent. More than a quarter of the people who use Muse engage in that activity, with an average purchase value above $200.
Zuckerberg also said Plaid reported that activity through its connector had not been seen on any other service. In the first week, Meta worked with Stripe to raise credit limits after transaction volume exceeded expectations.
The development raises questions about how AI-assisted shopping transactions will affect retailers and banks, especially if users go into debt and do not pay. Yahoo Finance raised bad debt and write-offs as concerns, not as losses already reported.
For companies involved in payments and sales, the scale of Muse activity does not by itself explain whether the transactions will generate healthy revenue. The raised credit limits show the companies responded to the surge in volume, but the source does not state the ultimate financial impact.
IBM spotlights returns on AI investment
IBM CEO Arvind Krishna said large companies are paying more attention to the real returns on their AI spending. He also said executives worry when spending on AI tools runs out of control without clear results.
“They are also getting worried if some of the AI spending completely out of control with no real return,” Krishna said on the earnings call. He maintained that the productivity and changes in ways of working that AI promises are real, in his view.
The three discussions show different yardsticks of AI success for each company: Tesla needs to train robots, Meta faces a surge of shopping activity through an agent, and IBM is responding to demands from companies that spending deliver benefits. The statements reported do not equate the issues, but they put development and cost control in the same conversation.
Attention now turns to whether Tesla can prepare Optimus for real-world tasks, how Meta manages Muse transactions after raising credit limits, and whether big companies see a commensurate return from their AI spending.
