Skip to main content
Live
Main content

Microsoft books $3.2B Anthropic gain, writes down OpenAI stake by $600M

One quarter of Anthropic paper gains nearly matched a full year of OpenAI markups for Microsoft's balance sheet.

Jaeden Schafer
Editor in Chief · · 4 min read
Microsoft logo

Microsoft booked a $3.2 billion paper gain on its Anthropic investment in the fourth quarter of fiscal 2026, adding 33 cents to diluted earnings per share of $4.81. In the same quarter, the company wrote down its OpenAI stake by roughly $600 million, subtracting about 7 cents from EPS. The split was pronounced enough that Microsoft called it out in its earnings disclosure.

The Anthropic gain traces back to Microsoft's $5 billion investment in the lab in November 2025, part of a circular arrangement in which Anthropic also committed to buying $30 billion worth of Azure services. Microsoft does not typically re-mark its Anthropic stake each quarter, which makes the $3.2 billion write-up a deliberate signal rather than a routine adjustment.

OpenAI gets different accounting treatment. Microsoft, which owns about 27% of OpenAI, updates that investment's carrying value every quarter and folds the change directly into earnings. Microsoft also receives revenue-share payments from OpenAI but does not break those out, so the equity mark is the visible number investors track.

Key facts

  • 01Microsoft booked a $3.2B gain on its Anthropic investment in Q4 FY2026, adding 33 cents to diluted EPS of $4.81.
  • 02The company marked down its OpenAI stake by $600M in the same quarter, subtracting 7 cents from EPS.
  • 03Microsoft invested $5B in Anthropic in November 2025 alongside a $30B Azure services commitment from the lab.
  • 04For the full fiscal year, the OpenAI stake generated a $5B gain and $0.67 of EPS on total EPS of $17.95.
  • 05Microsoft reported $90B in Q4 revenue and $35.8B in net income; full-year revenue was $331.8B.

For the fiscal year that ended June 30, the OpenAI position looks considerably better. It generated a $5 billion gain and contributed $0.67 to full-year EPS of $17.95. But the fact that a single quarter of Anthropic gains ($3.2 billion) nearly matched a full year of OpenAI gains ($5 billion) is the comparison Microsoft chose to highlight.

The $600 million OpenAI write-down is largely a rounding error against Microsoft's broader results. The company reported $90 billion in quarterly revenue and $35.8 billion in net income, with full-year revenue of $331.8 billion and net income of $133.7 billion. Neither AI mark meaningfully moves the top or bottom line.

What the disclosure does move is the narrative around Microsoft's dual-track AI strategy. Since the 2023 OpenAI board crisis, Microsoft has visibly diversified its lab exposure, first by hiring Mustafa Suleyman to run a consumer AI unit and then by writing the Anthropic check late last year. The Q4 numbers are the first quarter where the diversification shows up cleanly in accounting.

Anthropic's own trajectory helps explain the mark. The lab has been growing revenue rapidly on the back of Claude adoption in enterprise coding and agent workflows, and its most recent private-market valuations have run well above where Microsoft invested nine months ago. A $3.2 billion write-up on a $5 billion check implies Microsoft is carrying the stake at roughly 1.6x cost after less than a year.

OpenAI, by contrast, is in a more complicated financial phase. The company has been restructuring its capital arrangements with Microsoft and other investors, spending aggressively on compute buildouts, and navigating a shift toward a more conventional for-profit structure. A quarterly markdown against that backdrop is not a verdict on the product roadmap but a reflection of shifting inputs to Microsoft's valuation model.

Related · from this week
Palantir posts $1.9B quarter, Karp calls AI frontier labs 'Marxist'
Jaeden Schafer · 5 min read →

The obvious caveat is that both marks are unrealized. Microsoft cannot sell either stake into a public market, and the carrying values reflect internal assumptions about future cash flows, dilution, and comparable transactions. A quarter from now, the signs could flip. Microsoft has not disclosed the methodology it uses to arrive at either number.

The disclosure itself is the most interesting part. Microsoft did not have to draw attention to the divergence between its two AI bets, and the decision to spotlight the Anthropic gain suggests the company wants investors to see the second bet paying off. For an AI market that has spent two years treating OpenAI and Microsoft as functionally the same balance-sheet story, that is the shift worth watching.

ShareXLinkedInEmail
AI Box

Every AI model. One chat.

The latest models from ChatGPT, Claude, Gemini, Sora, ElevenLabs — 80+ models in a single chat. Compare answers side by side. Pick the best one every time.

  • ChatGPT, Claude, Gemini, Grok, DeepSeek — in one chat
  • Generate images & video with Sora, Veo, Ideogram
  • Compare any two models side by side
  • From $8.99/mo · 80+ models, all included
Try AI Boxaibox.ai
Trusted by 3,000+ teams
Got a tip?

Working on something we should cover, or seeing a story we missed? Send leads, documents, or feedback to hello@aichatdaily.com. For sensitive tips, see our secure tips page for Signal and PGP options.

Spotted an error? Email hello@aichatdaily.com with the URL and the issue, or read our full corrections policy.

AI Box Daily briefingFree · Daily · No fluff

Stay ahead of everyone in AI.

The tightly edited AI news email engineers, founders, and investors actually open. One email. Every weekday. Five minutes to finish.

Loved by 10,000+ AI professionals
Free forever. Unsubscribe with one click.

The briefing read inside teams at

Keep reading

More from Business

Palantir posts $1.9B quarter, Karp calls AI frontier labs 'Marxist'
Business

Palantir posts $1.9B quarter, Karp calls AI frontier labs 'Marxist'

Revenue jumped 93% year-over-year as CEO Alex Karp accused OpenAI and Anthropic of trying to capture their customers' means of production.

Jaeden Schafer5 min read
Microsoft logo
Business

Microsoft pitches its own AI models against OpenAI and Anthropic

Satya Nadella tells enterprises to stop trusting single frontier labs, pushing MAI models and Maya chips as cheaper swaps.

Jaeden Schafer5 min read
Microsoft logo
Business

Nadella warns AI buyers 'pay twice' — with cash and proprietary data

Microsoft's CEO says enterprises are teaching OpenAI and Anthropic their trade secrets, and pitches open source and cloud-hosted models as the fix.

Jaeden Schafer5 min read