AI Panic Gives Way to Company-by-Company Software Stock Sorting

Lauren Webster described the recent recovery in software shares as a turn away from the broad AI fear that drove a “massive sell-off” earlier in the year. The first phase of the market reaction, in her account, was uncertainty over what AI would do to traditional software companies and software stocks. By April, she said, software indices had produced their first constructive tape of the year — still below the broader market, but moving higher and performing “quite nicely.”

Her central point was not that AI anxiety has disappeared. It was that investors are becoming more selective. The broad fear that AI might impair incumbent software businesses is giving way to a more discriminating question: which companies can execute through the shift, which will benefit from AI, and which are more exposed to displacement.

That distinction matters because the same market can reward software shares on a rebound and still punish companies that look exposed to AI disruption, weak commercial demand, or insufficient profitability. Bloomberg’s lower-third set the discussion against investors bracing for more tech earnings; Webster’s framing was sharper: AI is moving from a generalized risk premium into a company-by-company test.

Caroline Hyde put Palantir at the center of the software-AI debate, describing it as “the eye of the storm” in questions about whether a software-exposed company can withstand competition from frontier model makers or chart its own path. The earnings frame captured the tension: “Palantir boosts outlook, misses commercial estimates.” Hyde added that Palantir’s numbers were “again extraordinary,” but the stock was down; Shopify was also lower.

Webster’s framework for judging durability started with customer base and deployment friction. Enterprise adoption, in her view, is materially different from small-business software adoption. Enterprise “rip and replace” is difficult: projects are budgeted years in advance, implementation timelines are long, and services are wrapped around the software. That makes it hard for customers to switch overnight, even if new AI-native tools appear.

By contrast, software that is easier to deploy can be changed faster. Webster said products that can be turned on more easily may be more vulnerable to customers adopting new technology from foundational AI labs. The distinction is not simply “AI company” versus “non-AI company,” but the operational embeddedness of a product: how deeply it is budgeted, implemented, serviced, and tied into a large organization’s workflows.

Her second filter was sector rotation, particularly into defense technology. Palantir’s relevance, in this framing, is not only its software exposure but its placement in a category drawing investor attention: defense tech companies with a software angle. Webster said there is “a real rotation” into the defense tech sector and expected continued tailwinds because of the geopolitical landscape, national-security spending, and “a new way of doing business” in defense technology.

Coinbase and PayPal brought the labor question into the same investor framework. Hyde cited Coinbase cutting 14% of workers and PayPal letting go thousands of workers, asking whether the AI explanation amounted to “AI washing” or reflected a broader fintech focus on efficiency.

Lauren Webster separated two forces that can coexist. First, she said investors are pushing companies harder toward profitability. The ratio of growth to profitability has changed over the past year, with profitability “mattering more than ever.” Companies looking for ways to make cuts that improve profitability, she said, would do so “regardless of AI.”

She did not dismiss AI’s labor impact. AI, in Webster’s view, is also going to displace certain types of tech workers. Her point was narrower than either extreme: not every announced cut should be understood as AI-driven, and AI is not irrelevant. The layoffs sit at the intersection of an investor preference for leaner, more profitable companies and a technological shift that can reduce demand for some categories of work.

The screen showed both fintech names trading higher intraday during the discussion.

Lauren Webster included a second-order effect in her labor view: AI adoption itself creates work. She said she is “really bullish” on the services sector that will emerge around AI adoption and implementation projects. The work she described is practical and enterprise-specific: figuring out how AI should be put in place, what security measures are needed, and how companies should manage implementation.

That point connects back to her moat framework. Enterprise software is hard to replace partly because implementation is complex, and AI implementation appears to have the same characteristic. Companies do not merely buy a model and switch it on across a large organization. They need services around deployment, security, and process design. Webster said that services push “requires some labor” as well.