WSJ Tech News Briefing

Tesla Reverses Course on Self-Driving Promise

14 min
Apr 24, 20264 days ago
Listen to Episode
Summary

The episode examines two major tech industry challenges: Salesforce's struggle to convince investors that AI will enhance rather than replace its enterprise software business, and Tesla's reversal on self-driving promises as customers with older hardware face exclusion from future autonomous capabilities. Both companies are pivoting their business models amid technological disruption and customer backlash.

Insights
  • Enterprise software companies face an existential threat from AI agents, but established players like Salesforce have early AI investments and compliance advantages that startups cannot easily replicate
  • Pricing model innovation (consumption-based vs. seat-based licensing) is critical for SaaS companies to capture value from AI-driven efficiency gains that reduce headcount
  • Tesla is repositioning itself from a car company to an autonomous software company, but this strategy depends on delivering promised self-driving capabilities that remain years away
  • Hardware obsolescence creates significant customer liability and legal risk when companies make long-term technology promises without clear upgrade paths
  • CEO credibility on timelines directly impacts investor confidence; Musk's history of overly optimistic predictions undermines Tesla's autonomy narrative
Trends
SaaSpocalypse: Enterprise software stocks declining as investors bet AI agents will replace traditional software licensing modelsShift from seat-based to consumption-based pricing in enterprise software to monetize AI capabilities rather than headcountHardware fragmentation creating customer segmentation in autonomous vehicle rollouts, with older hardware locked out of future capabilitiesBusiness model pivots from product sales to software-as-service positioning (Tesla emphasizing FSD software over vehicle sales)Global customer litigation campaigns against tech companies over unfulfilled autonomous/AI promisesAI startup competition threatening incumbent enterprise software vendors' market shareCompliance and regulatory complexity as competitive moat for established enterprise software playersCEO messaging shift: positioning AI as business opportunity rather than existential threat to legacy business models
Topics
AI Agent Adoption in Enterprise SoftwareSaaS Business Model DisruptionAutonomous Vehicle Technology PromisesHardware Obsolescence and Customer LiabilityFull Self-Driving (FSD) Regulatory and Legal ChallengesEnterprise Software Pricing Strategy EvolutionAI Investment and R&D in Legacy Tech CompaniesCustomer Class Action Litigation Against Tech CompaniesAutonomous Ride-Hailing Competition (RoboTaxi vs. Waymo)Humanoid Robotics CommercializationCEO Timeline Credibility and Investor ConfidenceCompliance as Competitive AdvantageAI Startup Threat to Enterprise SoftwareVehicle-as-Delivery-Mechanism StrategyInternational Customer Backlash on Tech Promises
Companies
Salesforce
Enterprise software company facing investor skepticism about AI strategy; CEO Mark Benioff argues AI will enhance rat...
Tesla
Facing lawsuits and customer backlash over unfulfilled self-driving promises; pivoting business model to emphasize au...
Anthropic
AI company that received early investment from Salesforce as part of its AI strategy
Waymo
Autonomous ride-hailing competitor to Tesla's planned RoboTaxi service
People
Mark Benioff
Argues Salesforce is positioned to capitalize on AI rather than be replaced by it; emphasizes early AI investments an...
Elon Musk
Announced on earnings call that Hardware 3 customers cannot access future autonomous driving; known for overly optimi...
Sebastian Herrera
Analyzed Salesforce's AI strategy and investor skepticism about enterprise software's future
Becky Peterson
Covered Tesla's self-driving promises, customer lawsuits, and business model pivot toward autonomous software
Imani Moise
Hosted Tech News Briefing episode covering Salesforce and Tesla stories
Tom Lesavio
Paid $8,000 for lifetime FSD access in 2019; now excluded from autonomy due to Hardware 3 obsolescence
Quotes
"Hardware 3 simply does not have the capability to achieve unsupervised FSD"
Elon MuskEarnings call, April 2024
"We have evolved our vehicle sales strategy, where we now emphasize FSD as a product and vehicle as only the delivery mechanism"
Tesla CFOEarnings call, April 2024
"Salesforce isn't really on the defense. It's building AI around all of its products"
Mark BenioffInterview segment
"You got us to pay thousands of dollars for a product that didn't exist and that still doesn't exist"
California class action plaintiffsLawsuit filing
"They have a whole suite of things for companies that you can't just easily replace. And they have a lot of compliance rules built around their products"
Sebastian HerreraInterview segment
Full Transcript
Welcome to Tech News Briefing. It's Friday, April 24th. I'm Imani Moise for The Wall Street Journal. Can software companies survive an AI apocalypse? Investors are skeptical, but one tech CEO says his company isn't being replaced by agents, it's cashing in on them. Then, Tesla has been fighting international backlash from car owners over its self-driving promises, while trying to convince Wall Street that it's more than just a car company. We're taking a look at how Elon Musk's priorities are shifting gears and what that means for customers. But first, it's been a tough time for software-as-a-service companies. Their stocks have been beaten down in recent weeks as investors place bets that the enterprise software that does everything from payroll to customer management will eventually be replaced by AI agents. Some are calling it the SaaSpocalypse, or the end of a business model that depends on selling software licenses per seat. But one person who's not worried? Mark Benioff, the CEO of enterprise software behemoth Salesforce. WSJ technology reporter Sebastian Herrera joins us now to explain why Benioff sees AI as a huge business opportunity for Salesforce and how he plans to prove investors wrong. Sebastian, Salesforce has become the poster child for the SaaSpocalypse. Its stock is down about 28% so far this year. What does Mark Benioff think markets are getting wrong? Yeah, he thinks that markets are really wrong for a few reasons. First of all, he says that Salesforce has been working on AI products since as early as 2014, when they created their AI research team and building up different AI tools within Salesforce, and even investing in AI companies like Anthropic pretty early on. And so what he's pitched is that Salesforce isn't really on the defense. It's building AI around all of its products. The flip side of it is just that a lot of investors feel like they haven't moved fast enough. They haven't seen enough proof that their AI products are really kind of moving the needle and for customers. And at the same time, you have all these AI startups who are building their own AI agents and their own automation that really threatens to sort of replace what Salesforce has done historically, which is traditional software. Now, what Mark Benioff says and what Salesforce says is that they have a whole suite of things for companies that you can't just easily replace. And they have a lot of compliance rules built around their products that you can't just vibe code your way into. And so far, what we've seen is that investors aren't really believing him fully yet, although that's been turning around a little bit recently. What do analysts say? Are they buying this argument? Analysts are starting to come around to the idea that this has been a little bit slow going for Salesforce, but everything that Salesforce is building around their products with AI is starting to work. They do have 23 customers signed up for their main AI products called Agent Force That a pretty small amount relative to the 150 customers that they serve But they did say recently that it's an $800 million business already. And you make a good point. Salesforce isn't new to AI. They already have some AI tools on the market. What are customers saying about those? When I talk to customers, they've liked how Salesforce's AI products, for example, help them to more automatically respond to customer inquiries about, for example, if you're selling books online and having a bot handle customer relationship questions, or if you're a sales representative and you're getting ready for a meeting with a client, there's agents from Salesforce that can help you prepare for those meetings. But at the same time, I've heard from customers that say that they would like more from Salesforce, that the agents that Salesforce have built can't handle complex emotional type of questions from customers. Like if someone's buying jewelry and is asking a question like, my wife likes pets, what should I get her? That agent force has struggled a bit with sort of those kinds of questions. Even if AI doesn't completely replace enterprise software needs, it could mean that companies need fewer employees. How will that impact Salesforce's business? Yeah, that's really the big challenge for Salesforce is that their customers just might become smaller companies. Then you might not need as many people to have licenses for Salesforce's software. Or you'll have smaller companies that are just trying out new things. And instead of going with a Salesforce, they might go with a hot new AI startup. So what Salesforce has done to adjust is they've changed the way they price their software, where you can still buy what's called a seat licenses for your employees, but you can purchase more on a consumption base for their AI agents. And so what they're hoping is that even if companies buy less seats for their employees, that they'll buy more of their agentic tools and capabilities and that that will be a growth sector for them. Maybe we'll get to a place where agents also need their own seats with companies like Salesforce. That's been some predictions by some people in the tech industry. So there's a lot to still to play out. And I think that's why you're seeing a lot of volatility still in the market is there's just so much uncertainty. That was WSJ technology reporter Sebastian Herrera. Are you replacing the software you use with AI? If you're a listener on Spotify, leave us a comment and let us know. Coming up, for some customers, Tesla's promises around self-driving have veered off course. That's after the break. For some Tesla owners, buying their vehicle was more than just purchasing a car. They also bought into the promise that their car would one day drive all by itself. And now some with cars equipped with an outdated technology called Hardware 3 feel they been taken for a ride Tesla is facing lawsuits and campaigns from different corners of the globe led by customers who paid thousands of dollars for pricey upgrades for a self product they claim still doesn exist For years, car owners and investors have repeatedly pressed the company for a clear timeline for when the cars would be capable of running the most sophisticated version of Tesla's driver-assist software. During an earnings call this week, Musk finally gave the update. WSJ Tesla reporter Becky Peterson is here to break down the story and what Tesla's latest earnings tell us about where they're going next. Becky, can you walk us through what some of these customers are saying? So there's a big lawsuit in California and there is a class action in Australia. And then some customers in Europe are organizing right now to figure out if they can take legal action. So in the California class action, which, by the way, Tesla is trying to get that class action status revoked. But that lawsuit is representing drivers in California who bought their cars inspired by some of the messaging from Tesla, which said that these cars will be able to be autonomous one day. And they're suing, saying this was false marketing. You got us to pay thousands of dollars for a product that didn't exist and that still doesn't exist. and they're trying to get refunds and they're also trying to get damages from Tesla for false advertising. How much did people pay for lifetime access to this technology? The price changed over time, but I spoke with this retired lawyer in California named Tom Lesavio and he bought his Tesla in 2019. Tesla has this product it calls FSD and it stands for full self-driving. He paid more than $100,000 in total for what was a luxury vehicle. And part of that was an $8,000 upgrade, which would give him access to the FSD technology for the lifetime of the car. But over time, Tesla developed new hardware. So Tom's car was basically outdated. Flash forward to today, most cars that are produced have something that's called hardware four. They're using the newest version of full self-driving, but Tom's car is basically too old to run the newest version of self-driving. And what has Tesla's response been? In the past, to make it right with customers who had older versions of the hardware, they gave free upgrades to both the computer and the cameras on the cars. That was around 2020, 2021. There's a lot of customers who've been asking Tesla, are you going to upgrade our cars or are we going to be able to have autonomy on our cars? Tesla has avoided giving specifics. But Wednesday on the earnings call, Musk said outright that customers with the old version of the hardware won't be able to use autonomy on their car. Hardware 3 simply does not have the capability to achieve unsupervised FSD They did clarify that they be offering either discounts on new cars or upgraded hardware to customers but they haven said when or where that's going to happen. One part of the problem is they haven't finished creating the autonomy software yet, so things could change. And on top of that, they're already starting to work with newer versions of computer chips. So there's a fifth and a sixth version that they're talking about now. So Tesla has emphasized the importance of full self-driving for the future of its business, particularly as it moves away from EVs to focus on autonomy, AI, and robotics. What did the company have to say about that on the earnings call? Its FSD product is one of the three main pillars of its future business. So they're going to be making these Optimus Humanoid robots. They're going to have an autonomous ride-hailing service, which is called RoboTaxi and which competes with Waymo. And they're going to offer this software to people who buy Tesla vehicles. Now, for the first time on the earnings call, the CFO said that they've changed their sales strategy. And now they're trying, when they sell cars, they're pitching it as a means of using the full self-driving software. With all this in mind, we have evolved our vehicle sales strategy, where we now emphasize FSD as a product and vehicle as only the delivery mechanism. So you may think that you need an electric vehicle to get from your house to work. But in fact, the electric vehicle is just a tool to let you use the autonomous driving software. Now, once again, customers don't have access to autonomous driving on their own cars. This is still a future-looking product. Musk said on the earnings call, maybe it will be available in Q4. But he's notorious for having these ambitious, overly optimistic timelines. And it usually happens a lot later than he claims. And what's at stake if Tesla doesn't get autonomous driving right? Well, they're losing their share of the car market, and they're not going to have a product to make money on. I mean, they're really taking a big risk, and there's always been promises. Maybe it will start making money on these things in 2026. So far, we're not seeing that. That was just their Q1 earnings, but Tesla is still a company that's very dependent on car sales. That was WSJ Tesla reporter Becky Peterson. And that's it for Tech News Briefing. If you're a listener on Spotify, be sure to leave us a comment. Today's show was produced by Julie Chang. I'm your host, Imani Moise. Additional support this week from Melanie Roy. Jessica Fenton and Michael LaValle wrote our theme music. Our supervising producer is Katie Ferguson. Our development producer is Aisha Al-Mouzling. And Chris Zinsley is deputy editor of audio for The Wall Street Journal. We'll be back later this morning with the TNB Tech Minute. Thanks for listening. Thank you.