Unlock Editor’s Digest for free
Roula Khalaf, editor-in-chief of the FT, selects her favourite stories in this weekly newsletter.
There’s one notable corner of the tech world that hasn’t yet been touched by the euphoria surrounding artificial intelligence in the stock market.
If generative AI really does represent the next big sales opportunity for the tech industry, software companies should be among the biggest winners. After all, most AI will likely appear as enhanced features in the enterprise software that companies rely on in their day-to-day operations.
But the BVP Nasdaq index of cloud software companies has fallen nearly 10 percent this year, while the Nasdaq Composite has risen more than 20 percent. It has also halved from its pandemic-era peak. The slump points to an industry at a crossroads. A long, secular growth phase fueled by the rise of the cloud appears to be entering a new, more mature state, while the next one—the spread of generative AI across the enterprise—has barely begun.
At times like these, Wall Street faces complex questions. If the cloud business is truly maturing, investors’ focus will need to shift more quickly from growth to value. Tech companies that have recently reported disappointing results, including Salesforce, MongoDB and Workday, have tried to disguise the calm as the result of persistent economic weakness. But the longer it goes on, the harder that argument becomes to sustain. Salesforce’s revenue has doubled over the past four years to $36 billion: at that scale, the slower 10 percent growth it has forecast for next year is starting to look more like the norm.
At the same time, investors must assess which companies will ride the next wave of growth and which will fail to adapt and be left behind.
According to the companies themselves, AI’s lack of impact on their sales is just a matter of timing. Salesforce CEO Marc Benioff, for example, points to the challenge of training large armies of salespeople to handle what he calls “a harder, more complex sale.” Customers are grappling with a wide range of questions, trying to understand how the new AI models work and how their employees should interact with them. They also have to consider how to redesign their workflows to take full advantage of the technology and deal with new threats to the security of their data.
Even if sales are still negligible, software companies are reporting huge customer interest in testing their new AI services. This could mean that the AI dividend has just been postponed.
Yet the disruptive threats of AI suggest it won’t be that simple. One is the upheaval in the business model of cloud companies. Most rely on per-seat subscriptions, meaning their revenues grow in line with the number of employees using their services. If generative AI works as promised and makes employees much more productive, customers should be able to do more with fewer staff.
The result is a shift toward consumption-based pricing, or charging based on how much the new services are actually used. Tying costs to usage has the added benefit of offsetting some of the higher costs of delivering generative AI. But unless this leads to real and demonstrable business benefits, software companies could face a backlash as customers see their bills rise.
Software groups also have a technical history. Historically, new technological eras—such as the rise of client-server computing in the 1990s and cloud computing in the decade that followed—have brought new waves of software start-ups to the forefront. New companies, their products and business models designed from the ground up to fit a new computing paradigm, start with a big advantage.
The first wave of these “AI native” software companies often seemed like little more than “wrappers” around the big language models, merely adding a layer of industry-specific expertise while offering companies ways to adopt generative AI. But they’re all working hard to gain a foothold from which they can start building more compelling services.
Salesforce’s Benioff says the incumbents will be tough to beat. Companies like his have become the repositories of their customers’ most important data, he says, which gives them a big advantage when it comes to training the AI models that companies will find truly useful.
That only matters if today’s cloud companies can adapt their own products and processes quickly enough to the new technology. Wall Street is reserving judgment for now.
richard.waters@ft.com