The headlines have been focused on Iran, but there’s no doubt that the coders and computer scientists at Anthropic, OpenAI, Alphabet, and scores of other artificial intelligence companies haven’t stepped away from their keyboards over the past week. Prior to the start of the war, stocks in one industry after another were hit by concerns that AI would undercut their business models.

Early in the AI boom, the emphasis was on the winners among the “picks and shovels” companies, like semiconductor or data center suppliers. But over the last five or six months, attention has shifted to the disruptive power of AI. In particular, investors are focused on how AI will lower barriers to entry across a wide swath of the service economy and breach economic moats among incumbent businesses big and small.

Economic moats, which can be thought of as a company’s competitive advantages, are a core element of how Morningstar’s equity research analysts evaluate stocks. Against the backdrop of the rapid advances in AI and resulting concerns across the markets, earlier this year, the equity research team cast a wide net across their coverage universe, reviewing the moat ratings of 132 tech and tech-adjacent companies. The results of this effort were rolled out late last week.

As of Friday, this review resulted in 37 moat downgrades and two upgrades. In addition 13 companies had the moat rating reaffirmed, but changes were made to fair value estimates or uncertainity ratings.

“The motivation for this was straightforward: The rapid advancement of large language models and agentic AI is likely bringing in a new technology paradigm that fundamentally alters the competitive dynamics for many software, services, and data businesses,” says Eric Compton, director of equity research, technology, for Morningstar.

“Traditional moat sources like switching costs and intangible assets remain relevant, but their durability now depends heavily on where a company sits in the technology stack, whether its data is truly proprietary, and whether its pricing model can withstand AI-driven changes in how enterprises consume software,” Compton says. “We felt it was no longer sufficient to assess moats without an explicit AI lens.”

This has been no small task. The team developed a seven-dimension AI disruption scoring framework which included data moat strength, pricing model resilience, switching costs, and current AI competitive position. Compton stresses that this rubric doesn’t replace the existing moat methodologies, but is instead meant to compliment them. “AI introduces new technology paradigms and how to analyze those moat sources, and also introduces more uncertainty into the system,” he says. “The point of this review was to make sure, as best we could, that we were updating our views and getting our calls right in light of the changing technology landscape.”

The list of stocks with downgrades includes Oracle, whose moat was downgraded to narrow from wide and whose fair value estimate was reduced to $215 per share from $277. Salesforce also had its moat reduced to narrow from wide and its fair value estimate to $280 per share from $300. Outside pure-play tech stocks, FactSet was downgraded to no moat from narrow, and Thomson Reuters was downgraded to a narrow moat from wide.

The limited moat upgrades include CrowdStrike, which went to a wide moat from narrow and had its fair value estimate upped to $460 per share from $410.

At the same time, our analysts kept their ratings on some 80 companies. Take Microsoft, which remains a wide-moat company. Morningstar senior equity analyst Dan Romanoff, who covers the stock, writes that “the firm should thrive regardless of AI.”

The story of AI and its effect on the markets and the global economy is clearly only beginning. Stay tuned this week as we take a deeper look at how Morningstar analysts are approaching the debate over AI’s impact on economic moats.

Get Morningstar insights in your inbox