Microsoft, the world’s largest software company and one of the earliest institutional backers of OpenAI, has become the worst-performing stock in the Magnificent 7 group of elite technology companies in 2026 — erasing roughly $857 billion in market value in the process.
The Context
The “Magnificent 7” — the loose grouping of Alphabet, Amazon, Apple, Meta, Microsoft, Nvidia, and Tesla — has functioned as shorthand for the AI investment thesis throughout the current technology cycle. As a category, these stocks have attracted enormous institutional capital on the premise that scale, compute access, and data moats would translate into durable AI-era earnings power. For most of the group, that thesis has held up reasonably well. For Microsoft, it has quietly begun to unravel.
Microsoft’s involvement in AI is not superficial. The company has committed billions in capital expenditure to cloud infrastructure, embedded its Copilot AI assistant across its enterprise software suite, and deepened its strategic partnership with OpenAI. By most technical and strategic measures, Microsoft looked like one of the best-positioned incumbents to profit from the AI wave. The market, at least for now, disagrees.
To understand why, it helps to understand the dual pressures now converging on software-first companies that have made large AI commitments — a dynamic that the AI market’s broader inflection has made considerably sharper in 2026.
The Move
Microsoft’s stock is on pace for an 18% decline in June alone — its worst single month since the dot-com bust of 2000. Year to date, shares are down approximately 24%, the steepest fall among all Magnificent 7 constituents. The stock is now trading near levels last seen in 2023, and its forward price-to-earnings ratio fell to roughly 21 times as recently as last week — a multi-year low for a company that has historically commanded a significant valuation premium.
That P/E compression is significant. It signals that investors are not simply trimming a richly valued position; they are genuinely re-rating the company’s earnings outlook. The question now dividing analysts and traders is whether 21x forward earnings represents a fair reassessment of Microsoft’s AI-era prospects, or an overreaction that has opened a buying opportunity.
Michael Burry — the investor immortalized by his prescient bet against mortgage securities ahead of the 2008 financial crisis — publicly declared himself in the buyer’s camp. In a Substack post published last Thursday, Burry disclosed that he had purchased call options on Microsoft stock structured to pay off if shares rise into the low $700s by 2028. His post alone was sufficient to spark a 6% single-day rally in the stock the following Friday, a reminder of how much retail sentiment can still move even a multi-trillion-dollar name when institutional conviction is thin.
The Stakeholders
Microsoft
The company’s predicament is what market observers are calling a “double whammy.” Microsoft faces simultaneous investor scepticism on two entirely separate fronts: the first is the scrutiny directed at any large-cap company spending heavily on AI infrastructure without near-term earnings payback; the second — and arguably more structurally threatening — is the fear that AI itself will erode the value of Microsoft’s core software franchise.
This second concern is the one that distinguishes Microsoft from, say, Nvidia or Alphabet. Microsoft is, at its foundation, a software company. Its Office suite, its Windows operating system, its enterprise productivity tools — these are products that generate reliable, high-margin recurring revenue. AI-native competitors and AI-assisted coding tools now pose a credible, if still early-stage, challenge to that model. The irony is stark: Microsoft’s own AI investments may be accelerating the disruption of its own most profitable business lines. As Blockgeni has reported, software developers are increasingly experimenting with AI-assisted coding in ways that compress the value of traditional software tooling.
What makes Microsoft’s position uniquely difficult — and what the source reporting does not fully surface — is that its diversification, historically its greatest defensive advantage, has become a liability in the current environment. Precisely because Microsoft operates across software, cloud, gaming, and enterprise services, it carries maximum exposure to AI disruption across multiple business lines simultaneously. A company like Nvidia has no legacy software revenue to cannibalise. A company like Meta operates in a single, clearly defined ecosystem. Microsoft, by contrast, faces a kind of distributed disruption risk: AI threatens something meaningful in almost every segment it operates. Diversification shields against concentrated shocks, but it offers little protection when the shock is systemic.
Oracle
Microsoft is not alone in this particular no-man’s-land. Oracle, another legacy enterprise software giant that has made aggressive moves into cloud and AI infrastructure, is experiencing a strikingly similar stock chart in 2026. Like Microsoft, Oracle has committed heavily to AI capital expenditure. Like Microsoft, it is facing investor impatience over the timeline to returns. And like Microsoft, it sits awkwardly between the old software economy and the new AI infrastructure one.
The key structural difference between the two companies is how that capital expenditure is being financed. Oracle has leaned heavily on debt to fund its AI buildout, which has placed its bond market performance under unusual scrutiny. Oracle’s corporate debt has, in some quarters, begun to function as a proxy indicator — a bellwether, in the terminology of fixed-income traders — for investor sentiment toward the broader AI infrastructure trade. When confidence in AI spending timelines wavers, Oracle’s bonds move. That dynamic adds a layer of financial market complexity that Microsoft, which carries a considerably stronger balance sheet, does not face to the same degree.
Michael Burry and Retail Sentiment
Burry’s disclosure deserves its own analysis. His call options are structured around a multi-year timeline — a 2028 expiry — which is itself a statement of conviction about how long the current uncertainty may persist. He is not betting on a quick reversal; he is betting that the market’s current pessimism about Microsoft’s AI trajectory will, over two years, prove excessive. Whether or not his thesis is correct, his public disclosure has already demonstrated something important: retail and semi-institutional investors are watching for any credible signal that the selloff has gone too far, and they will respond quickly when one arrives. The 6% single-session rally on the back of a Substack post is a measure of how fragile the bearish consensus actually is.
How Microsoft Compares to Other Magnificent 7 Members Under AI Pressure
Microsoft’s weak 2026 performance compared with other major AI-linked technology companies reflects a broader market reassessment of how AI affects legacy software businesses. The companies under the most pressure, including Microsoft and Oracle, share a common problem: they must spend heavily on AI infrastructure while also defending software and services revenue that AI could eventually commoditize. This creates a “double whammy” that companies with cleaner AI infrastructure exposure, such as Nvidia, have largely avoided. Apple faces concern over its AI lag, Meta is under pressure because AI is central to its growth story, and Alphabet faces search disruption risk, but Microsoft’s challenge is more complex because AI threatens both its cost structure and parts of its core enterprise software model. Investor psychology is also being shaped by fears that AI could reduce demand for software jobs and, by extension, the tools those workers use. Whether that fear is overstated remains uncertain, but the market is clearly no longer assuming that scale, distribution, and legacy dominance will automatically translate into AI-era earnings.
How Serious Players Should Respond
For institutional investors and portfolio managers, the Microsoft situation demands a clear-eyed distinction between valuation compression driven by genuine structural risk and compression driven by narrative overcorrection. A forward P/E of 21x for a company with Microsoft’s balance sheet strength, cloud infrastructure scale, and enterprise customer lock-in is not obviously justified by the fundamentals — but it is not obviously wrong either. The right response is rigorous scenario analysis, not reflexive contrarianism. Michael Burry’s 2028-dated call options represent one version of a structured, time-bounded bet on mean reversion; institutions without his risk tolerance and timeline flexibility should approach the position with corresponding caution.
For corporate executives at software-first companies — whether at Microsoft itself or at the broader ecosystem of enterprise software vendors now facing similar questions — the strategic imperative is to move faster on demonstrating concrete AI-driven revenue, not just AI-driven cost efficiency. The market is not currently rewarding AI investment as a promise; it is demanding evidence that the investment is already translating into durable, defensible new revenue streams. Companies that can show customer adoption data, retention improvements, or measurable productivity gains attributable to their AI products will be better positioned to escape the valuation trap that is currently snaring Microsoft.
For regulators and policymakers watching the technology sector, the Microsoft episode offers a useful data point: markets are beginning to self-correct some of the most exuberant AI investment assumptions without regulatory intervention. That is broadly healthy. But the concentration of AI infrastructure investment in a small number of heavily scrutinized public companies — and the debt-financed exposure of firms like Oracle — warrants continued attention to systemic financial risk as the AI capital expenditure cycle matures. The broader reshaping of capital markets by technology-driven investment cycles is a pattern regulators would do well to monitor closely, even when the immediate crisis appears to be resolving itself.
