The numbers are stated judgments. The record began on 15 October 2026 with starting confidences that a model set as the average of five separate readings, from its general knowledge and up to ten developments for each assessment first reported in the three weeks before, after a second model had checked the developments the readings relied on, and a published rule has made every change since. They are not measured probabilities. The question and the rule are fixed before the record begins, and each change shows its evidence and its reasons. The rule and what is not claimed are on the method page.

This record corrects the first record of the same question, Microsoft, which began on 13 October 2026 and continues unchanged under the rules it began with. What changed, and why is on this record’s method page.

Main assessment

Microsoft's revenue from AI grows fast enough to cover the rising cost of the capacity it builds to serve it, so that through the end of 2027 Microsoft's operating profit keeps growing and Microsoft does not cut back its capital spending for lack of demand or returns.

67.5% stated confidence, Likely (55–80%). Starting point on the record’s first day: Likely (55–80%), 67.5 per cent, the average of five readings, all five of Likely. Attention level: For information. Trend: Steady.

Stated confidence in the one daily state so far: 15 October 2026, 67.5 per cent. A chart is drawn once there are 3 daily states.

Key judgment

The judgment as stated before the record began

Each quarter Microsoft spends more on new capacity, much of it on chips and servers that are written off over a few years, so the cost of AI arrives faster than the revenue it brings. This assessment holds that the revenue keeps up. In the fiscal year to June 2026 Microsoft's operating income rose 21 per cent, customer demand exceeded the capacity it had available, and the revenue its commercial customers had committed to under contract but Microsoft had not yet recognized rose 84 per cent in a year, to US$678 billion. If revenue keeps pace, operating profit keeps growing while spending rises, and Microsoft has no reason to cut back the capacity it builds. The assessment would weaken if operating profit fell from a year earlier while spending kept rising, if Microsoft cut its spending outlook and cited demand or returns, or if the AI model developers that account for a large share of its contracted business paid less, later or elsewhere.

Starting point

How the record started this assessment

On 15 October 2026, before the record began, a model read the 10 developments below five separate times, first reported in the three weeks before and linked to this assessment's causal factors or to its rivals', and chose a term each time. A second model, from a different family, checked each development a reading relied on, disputed none, and the five readings were made again. The record started the assessment at the average of the five final terms, taken in log-odds: 67.5 per cent, Likely (55–80%). Every reading and every check is kept in the first day's record.

The five final readings, in order: Likely, Likely, Likely, Likely, Likely. Their average, taken in log-odds, is 0.73, a stated confidence of 67.5 per cent. The readings were made by Claude Sonnet 4.6 (Anthropic) and checked by Gemini 4 Argon (Google).

Factors named above: M1 (Operating profit growing alongside spending), M2 (Demand for capacity running ahead of supply), M3 (What businesses pay for AI), M4 (A falling cost for each unit of AI work), M5 (AI model developers paying less, later or elsewhere), C3 (Higher prices for chips, memory and power) and C4 (AI model developers paying less, later or elsewhere).

The developments shown

1. First reported 14 October 2026, 1 independent publisher. Linked to this assessment's own causal factors.
“Microsoft Disputes Azure 20% Cost-Cut Report as Maia 300 AI Expansion Looms” · windowsforum.com · 14 October 2026 · Open the report · archived copy
First round: relied on by 3 readings as support for the assessment. Second round: relied on by 3 readings as support for the assessment.
- Check of its use as support for the assessment: Upheld by the second model, run on Gemini 4 Argon (Google).
  Its reason, as the model wrote it: Report 1 directly states that Microsoft disputed an Azure cost-cut report as its Maia 300 AI chip expansion looms to help manage infrastructure costs for Copilot and other AI workloads. Deploying custom AI chips to manage infrastructure costs while maintaining investment supports the assessment and factor M4 by pointing toward lower unit costs for AI work without retreating on capacity.
  - Report 1, quoted: “Microsoft Disputes Azure 20% Cost-Cut Report as Maia 300 AI Expansion Looms”
  - Report 1, quoted: “The report describes Microsoft’s denial of an Azure cost-cutting directive alongside plans to deploy Maia chips for Copilot and other AI workloads, amid pressure to manage infrastructure costs.”
2. First reported 13 October 2026, 1 independent publisher. Linked to this assessment's own causal factors.
“Microsoft now has 3 generations of its own AI chips. Nvidia is still the chip stock to buy.” · finance.yahoo.com · 13 October 2026 · Open the report · archived copy
First round: relied on by 5 readings as support for the assessment. Second round: relied on by 5 readings as support for the assessment.
- Check of its use as support for the assessment: Upheld by the second model, run on Gemini 4 Argon (Google).
  Its reason, as the model wrote it: Report 1 confirms that Microsoft has developed three generations of its own AI chips and notes that the Maia 300 could help shift more inference workloads to its own silicon. Advancing proprietary chips to handle AI workloads directly relates to Factor M4 (falling cost per unit of AI work through more efficient chips) and supports the main assessment that AI revenue will cover capacity costs.
  - Report 1, quoted: “Microsoft now has 3 generations of its own AI chips.”
  - Report 1, quoted: “Maia 300 could help Microsoft shift more inference workloads to its own chips”
3. First reported 13 October 2026, 1 independent publisher. Linked to this assessment's own causal factors.
“Microsoft unveils Maia 300 chip, plans gigawatt-scale AI buildout” · en.sedaily.com · 13 October 2026 · Open the report · archived copy
First round: relied on by 5 readings as support for the assessment. Second round: relied on by 4 readings as support for the assessment.
- Check of its use as support for the assessment: Upheld by the second model, run on Gemini 4 Argon (Google).
  Its reason, as the model wrote it: The report explicitly states that Microsoft unveiled its custom Maia 300 AI accelerator and plans a gigawatt-scale buildout starting in 2027. Deploying custom silicon at scale is a step toward more efficient chips and data centers to lower the unit cost of AI work, which directly supports factor M4 and the main assessment. None of the four grounds for challenge apply.
  - Report 1, quoted: “Microsoft unveils Maia 300 chip, plans gigawatt-scale AI buildout”
  - Report 1, quoted: “Microsoft announced its Maia 300 AI accelerator and plans to begin deployment in 2027, advancing its strategy to build large-scale AI infrastructure using custom silicon.”
4. First reported 12 October 2026, 1 independent publisher. Linked to a rival's causal factors.
“Intel Accelerates Open Infrastructure for the Agentic AI Era” · finance.yahoo.com · 12 October 2026 · Open the report · archived copy
First round: relied on by no reading. Second round: relied on by no reading.
No reading relied on it, so it was not checked.
5. First reported 12 October 2026, 1 independent publisher. Linked to this assessment's own causal factors.
“d-Matrix Announces Squadron Compute Clusters for AI Inference at Data Center Scale” · finance.yahoo.com · 12 October 2026 · Open the report · archived copy
First round: relied on by no reading. Second round: relied on by no reading.
No reading relied on it, so it was not checked.
6. First reported 12 October 2026, 1 independent publisher. Linked to this assessment's own causal factors.
“Microsoft Needs Azure to Keep Growing Faster Than 40% When It Reports on Oct. 28” · finance.yahoo.com · 12 October 2026 · Open the report · archived copy
First round: relied on by 2 readings as support for the assessment. Second round: relied on by 4 readings as support for the assessment.
- Check of its use as support for the assessment: Upheld by the second model, run on Gemini 4 Argon (Google).
  Its reason, as the model wrote it: Report 1 explicitly notes 'persistent demand exceeding available capacity, making data-center buildout a key constraint on growth' ahead of Microsoft's Oct. 28 earnings report. This directly matches factor M2 (customer demand for Microsoft's AI and cloud capacity exceeding what it has available) and supports the main assessment over alternatives where AI revenue fails to cover capacity costs.
  - Report 1, quoted: “Microsoft Needs Azure to Keep Growing Faster Than 40% When It Reports on Oct. 28”
  - Report 1, quoted: “persistent demand exceeding available capacity, making data-center buildout a key constraint on growth”
7. First reported 9 October 2026, 1 independent publisher. Linked to this assessment's own causal factors.
“Who Actually Pays for the AI Data Centers Built for OpenAI and Anthropic” · coinpaper.com · 9 October 2026 · Open the report · archived copy
First round: relied on by no reading. Second round: relied on by 1 reading as support for the assessment and 1 reading as counting against it.
No reading relied on it, so it was not checked.
8. First reported 8 October 2026, 1 independent publisher. Linked to this assessment's own causal factors.
“Cloud Giants Are Building Their Own AI Chips. The Numbers Explain Why — DutyStation News” · dutystation.ai · 8 October 2026 · Open the report · archived copy
First round: relied on by 2 readings as support for the assessment. Second round: relied on by 2 readings as support for the assessment and 1 reading as counting against it.
- Check of its use as support for the assessment: Upheld by the second model, run on Gemini 4 Argon (Google).
  Its reason, as the model wrote it: The report states that Microsoft has brought its homegrown Maia accelerators into its datacenters and that the economics of scale drive cloud giants to build their own AI chips. This directly supports factor M4 (a falling cost for each unit of AI work through more efficient chips), which in turn supports the main assessment that Microsoft's AI revenue will cover the cost of capacity.
  - Report 1, quoted: “Cloud Giants Are Building Their Own AI Chips. The Numbers Explain Why”
  - Report 1, quoted: “AWS, Google, and Microsoft train and serve on homegrown silicon alongside Nvidia. The economics of scale - not nationalism - drive the shift”
  - Report 1, quoted: “Microsoft has brought its Maia accelerators into its datacenters.”
9. First reported 1 October 2026, 1 independent publisher. Linked to a rival's causal factors.
“Microsoft and OpenAI Rewrite Their AI Alliance: Freedom, Caps and a Looser Bond” · webpronews.com · 1 October 2026 · Open the report · archived copy
First round: relied on by 1 reading as counting against it. Second round: relied on by 2 readings as counting against it.
- Check of its use as counting against the assessment: Upheld by the second model, run on Gemini 4 Argon (Google).
  Its reason, as the model wrote it: Report 1 states that Microsoft and OpenAI have amended their partnership by ending exclusivity and giving OpenAI the freedom to sell across clouds. This directly supports factor C4—an AI model developer renegotiating commitments or moving work to other cloud providers—which counts in favor of assessment if_wrong and therefore against the main assessment.
  - Report 1, quoted: “Microsoft and OpenAI Rewrite Their AI Alliance: Freedom, Caps and a Looser Bond”
  - Report 1, quoted: “Microsoft and OpenAI have amended their landmark partnership twice since late 2025, ending exclusivity, capping revenue shares and removing the AGI trigger. The changes give OpenAI freedom to sell across clouds”
10. First reported 29 September 2026, 1 independent publisher. Linked to this assessment's own causal factors.
“TSMC is raising 3nm chip prices up to 15% and nobody can say no - Startup Fortune” · startupfortune.com · 29 September 2026 · Open the report · archived copy
First round: relied on by 5 readings as counting against it. Second round: relied on by 4 readings as counting against it.
- Check of its use as counting against the assessment: Upheld by the second model, run on Gemini 4 Argon (Google).
  Its reason, as the model wrote it: Report 1 states that TSMC is raising prices for 3nm chips, which include the leading-edge chips used in AI. Under the definitions of factors M4 and C3, rising prices for AI chips and components raise the cost of capacity and work against falling unit costs, thereby counting against the main assessment.
  - Report 1, quoted: “TSMC is raising 3nm chip prices up to 15% and nobody can say no”
  - Report 1, quoted: “TSMC makes essentially every leading-edge chip that matters in AI right now, and it just told its biggest customers to pay more for the privilege.”

The readings, and their reasons

The first round, before the check

Reading 1: Likely (55–80%), from 10 developments shown. It relied on development 3 as support for the assessment, development 2 as support for the assessment, development 8 as support for the assessment, development 1 as support for the assessment, development 10 as counting against the assessment.
Its reason, as the model wrote it: Development 3 and development 2 show Microsoft advancing its own chip program with the Maia 300 and planning a gigawatt-scale buildout, which supports factor M4 by pointing toward falling unit costs over time. Development 8 explains why cloud giants build their own chips, reinforcing that logic. Development 1 shows Microsoft disputing a cost-cut report while the Maia 300 expansion looms, which supports M2 by suggesting demand remains strong enough that Microsoft is not retreating. Development 10 points to rising TSMC chip prices, which works against M4. Before these developments, Microsoft had already been reporting growing operating income alongside rising capital spending, which is the core of factor M1, and Azure demand has consistently been described as exceeding supply. On balance the prior evidence and the direction of most developments support this assessment.
Reading 2: Likely (55–80%), from 10 developments shown. It relied on development 3 as support for the assessment, development 2 as support for the assessment, development 6 as support for the assessment, development 10 as counting against the assessment.
Its reason, as the model wrote it: Development 3 reports Microsoft unveiling the Maia 300 chip and planning gigawatt-scale AI buildout, suggesting demand justifies continued heavy investment and supports M2. Development 2 confirms Microsoft now has three generations of its own AI chips, which supports M4 by pointing toward falling unit costs over time. Development 6 notes market pressure for Azure to keep growing faster than 40 percent, which is consistent with demand running ahead of supply and supports M2. Development 10, linked through the rival, reports TSMC raising 3nm prices up to 15 percent, which modestly undercuts M4. Before these developments, Microsoft had been reporting rising operating profit alongside rising capital spending for several consecutive quarters, which is the core condition M1 describes.
Reading 3: Likely (55–80%), from 10 developments shown. It relied on development 3 as support for the assessment, development 2 as support for the assessment, development 6 as support for the assessment, development 10 as counting against the assessment, development 9 as counting against the assessment.
Its reason, as the model wrote it: Development 3 reports Microsoft unveiling its Maia 300 chip and planning gigawatt-scale AI buildout, suggesting strong demand signals supporting continued capital commitment rather than cutbacks. Development 2 confirms Microsoft has three generations of its own AI chips, which supports factor M4 by indicating falling unit costs through in-house silicon. Development 6 notes market expectations for Azure growth above 40 percent, reflecting broad analyst confidence in revenue covering costs. Development 10, reporting TSMC raising 3nm prices up to 15 percent, counts against the assessment by raising input costs. On balance, Microsoft's demonstrated operating profit trajectory and custom chip progress support this assessment, though chip cost pressures and the OpenAI relationship renegotiation in development 9 introduce meaningful uncertainty.
Reading 4: Likely (55–80%), from 10 developments shown. It relied on development 2 as support for the assessment, development 3 as support for the assessment, development 1 as support for the assessment, development 8 as support for the assessment, development 10 as counting against the assessment.
Its reason, as the model wrote it: Developments 2 and 3 report Microsoft unveiling its Maia 300 chip and planning gigawatt-scale buildout, pointing to falling unit costs over time which supports M4. Development 1 reports Microsoft disputing a 20 percent Azure cost-cut claim, suggesting it believes its cost position is manageable, also supporting M4. Development 8 reports cloud giants building their own chips precisely because the economics improve margins, further supporting M4. Development 10 reports TSMC raising 3nm prices up to 15 percent, which counts against M4 for near-term costs. On balance, the pre-existing record of growing operating profit alongside rising capex, combined with the in-period signals on proprietary chips and gigawatt expansion, keeps the assessment likely.
Reading 5: Likely (55–80%), from 10 developments shown. It relied on development 2 as support for the assessment, development 3 as support for the assessment, development 1 as support for the assessment, development 10 as counting against the assessment.
Its reason, as the model wrote it: Development 2 reports Microsoft now has three generations of its own AI chips, which supports factor M4 by showing falling unit costs through in-house silicon. Development 3 reports Microsoft is planning a gigawatt-scale AI buildout alongside the Maia 300 launch, which supports M2 by signalling demand justifies continued large capacity expansion. Development 1 reports Microsoft disputed a claim that it was cutting Azure costs by 20 per cent, which weakens any reading that revenue is falling short. Development 10 reports TSMC is raising 3nm chip prices up to 15 per cent, which supports M4's opposite direction and mildly undercuts the assessment. Before these developments, Microsoft's sustained operating income growth alongside rising capital spending was widely known and supports the assessment.

The second round, after the check, which set the start

Reading 1: Likely (55–80%), from 10 developments shown. It relied on development 2 as support for the assessment, development 8 as counting against the assessment, development 1 as support for the assessment, development 6 as support for the assessment, development 9 as counting against the assessment.
Its reason, as the model wrote it: Development 2 and development 8 together describe Microsoft building its own AI chips across three generations while TSMC raises 3nm prices up to 15 percent, pointing in opposite directions on unit cost. Development 1 reports Microsoft disputing a claim of a 20 percent Azure cost cut while simultaneously announcing the Maia 300 expansion, which suggests demand remains strong enough to justify continued buildout. Development 6 notes that analysts expect Azure to keep growing faster than 40 percent, supporting the view that revenue is covering capacity costs. Development 9, linked through the rival, describes a looser OpenAI arrangement that introduces some uncertainty about a major customer, but the broader picture of ongoing investment and strong Azure growth expectations leaves the assessment more likely than not.
Reading 2: Likely (55–80%), from 10 developments shown. It relied on development 3 as support for the assessment, development 2 as support for the assessment, development 1 as support for the assessment, development 6 as support for the assessment, development 10 as counting against the assessment, development 7 as counting against the assessment.
Its reason, as the model wrote it: Development 3 reports Microsoft unveiling its Maia 300 chip and planning gigawatt-scale buildout, and development 2 confirms Microsoft now has three generations of its own AI chips, both supporting factor M4 that falling unit costs help revenue cover capacity. Development 1 reports Microsoft disputing an Azure cost-cut report while the Maia 300 expansion looms, which supports factor M2 that demand remains strong. Development 6 notes Azure must keep growing faster than 40 percent, which is a high bar but reflects analyst expectations of continued growth consistent with M1. Development 10 reports TSMC raising 3nm chip prices up to 15 percent, which is a modest count against via factor M4. Development 7 touches on who pays for AI data centers built for OpenAI and Anthropic, linked through the rival, suggesting some concentration risk but not a clear reversal. On balance, widely known prior evidence of strong Azure momentum and operating profit growth alongside rising capex tilts the assessment to likely.
Reading 3: Likely (55–80%), from 10 developments shown. It relied on development 3 as support for the assessment, development 8 as support for the assessment, development 2 as support for the assessment, development 6 as support for the assessment, development 10 as counting against the assessment, development 7 as support for the assessment.
Its reason, as the model wrote it: Development 3 reports Microsoft unveiling its Maia 300 chip and planning gigawatt-scale AI buildout, supporting M4 as falling unit costs help revenue cover capacity. Development 8 explains why cloud giants build their own chips, again supporting M4. Development 2 confirms Microsoft now has three generations of its own AI chips, supporting M4. Development 6 notes Azure must keep growing faster than 40 percent, reflecting strong demand consistent with M2. Development 10, TSMC raising 3nm prices up to 15 percent, counts against M4 by raising input costs. Development 7, reporting on who pays for AI data centers built for OpenAI and Anthropic, is mixed but suggests Microsoft's capacity commitments are backed by contractual obligations. Prior knowledge that Microsoft's operating profit has grown steadily alongside rising capital spending provides a strong baseline for this assessment.
Reading 4: Likely (55–80%), from 10 developments shown. It relied on development 8 as support for the assessment, development 2 as support for the assessment, development 3 as support for the assessment, development 6 as support for the assessment, development 10 as counting against the assessment, development 9 as counting against the assessment.
Its reason, as the model wrote it: Development 8 reports cloud giants building their own AI chips to reduce unit costs, and developments 2 and 3 show Microsoft now fielding three generations of its own Maia chips with a gigawatt-scale buildout planned, both supporting factor M4 on falling unit costs. Development 6 notes analyst expectation of Azure growth above 40 percent, supporting M1 and M2 on demand running ahead of supply and operating profit growing alongside spending. Development 10 on TSMC raising 3nm prices up to 15 percent counts against the assessment through factor M4, but Microsoft's own-chip push partially offsets that pressure. Development 9, linked through the rival, reports a looser OpenAI bond with caps, which touches M5, though it is a single report and prior knowledge of Microsoft's broad enterprise base limits its weight. On balance the evidence and what was widely known about Microsoft's growing enterprise AI revenue support a likely judgment.
Reading 5: Likely (55–80%), from 10 developments shown. It relied on development 2 as support for the assessment, development 3 as support for the assessment, development 1 as support for the assessment, development 10 as counting against the assessment.
Its reason, as the model wrote it: Development 2 reports Microsoft now has three generations of its own AI chips, which supports factor M4 by showing falling unit costs through in-house silicon. Development 3 confirms the Maia 300 expansion and gigawatt-scale buildout, reinforcing that capacity is being matched to expected demand and supporting M2. Development 1 shows Microsoft disputing a report of a 20 percent Azure cost cut, which if anything signals continued pricing confidence and supports M3. Development 10 reports TSMC raising 3nm prices up to 15 percent, which mildly undercuts M4, but Microsoft's shift to its own chips partially offsets this. Before these developments Microsoft had sustained growing operating profit alongside rising capital spending, and the developments modestly reinforce that picture.

Readings made before this one

No starting point was read before this one, as the first day records.

Uncertainty

Causes of uncertainty

Microsoft does not report its revenue from AI as a regular separate figure, so whether that revenue covers the cost of AI capacity has to be read from operating income, revenue under contract and the gross margin on the Microsoft Cloud, which all include other business.
Microsoft raised list prices for many of its business software suites from July 2026 and pointed to added AI, security and management features, so revenue per user can rise without any new use of AI.
A large share of the revenue Microsoft has under contract comes from a few AI model developers, whose ability to pay over many years depends on money they have yet to raise.
From July 2026 Microsoft treats its data centre buildings as lasting 25 years in place of 15, which slightly lowers depreciation and so slightly raises operating profit without any change in what AI earns.

What would change it

What would change the judgment

Microsoft's operating profit falling from a year earlier in two quarters running while its capital spending keeps rising.
Microsoft lowering its capital spending outlook and citing demand or returns.
Microsoft saying that it has capacity to spare while the revenue it has under contract stops growing.

Causal factors

The causal factors, and which way each one cuts

Factor	Which way it cuts	What evidence looks like	Last moved
M1. Operating profit growing alongside spending Microsoft's operating profit, its operating income, growing year over year while its capital spending rises. This shows revenue covering the rising cost of the capacity, which supports this assessment.	Supports. A stronger factor supports the assessment.	An earnings release shows operating income up from a year earlier while capital spending rises. Microsoft keeps its outlook for operating margin close to the previous year's. An earnings release shows operating income lower than a year earlier, which bears on the factor in the other direction.	Not yet
M2. Demand for capacity running ahead of supply Customer demand for Microsoft's AI and cloud capacity exceeding what it has available, so that new capacity earns revenue as soon as it is ready. Demand that runs ahead of supply supports this assessment. Capacity left unused for lack of customers weakens it.	Supports. A stronger factor supports the assessment.	Microsoft says on its earnings call that customer demand exceeds the capacity it has available. Microsoft reports that Azure growth was held back by the capacity it could bring into use. Microsoft says that it has capacity to spare, which bears on the factor in the other direction.	Not yet
M3. What businesses pay for AI Whether businesses accept higher list prices or new paid tiers for AI in Microsoft's business software. Prices that hold show that businesses see enough value to keep paying, and let revenue keep pace with the cost of the capacity behind it.	Supports. A stronger factor supports the assessment.	Microsoft raises list prices or adds a paid tier for AI in its business software. Microsoft reports revenue per user rising in its business cloud and links it to AI. Microsoft cuts prices, deepens discounts or includes AI at no extra charge to widen use, which bears on the factor in the other direction.	Not yet
M4. A falling cost for each unit of AI work The cost to Microsoft of each unit of AI work it delivers falling, through more efficient chips, software and data centres, faster than prices rise for the chips, memory and power that AI capacity needs. A falling unit cost lets revenue cover the capacity more easily, which supports this assessment.	Supports. A stronger factor supports the assessment.	Microsoft reports delivering more AI work from the same chips or the same power. Microsoft reports a lower cost for each unit of AI work it delivers. Makers of memory chips or AI accelerators raise their prices, which bears on the factor in the other direction.	Not yet
M5. AI model developers paying less, later or elsewhere AI model developers that have committed to Azure capacity reducing, delaying or renegotiating those commitments, or moving work to other providers, which leaves capacity built for them earning less than planned. More of this undercuts this assessment.	Weakens. A stronger factor weakens the assessment.	Microsoft reports a reduction or renegotiation of a large commitment from an AI model developer. An AI model developer reports difficulty raising the money to pay for computing it has contracted. An AI model developer moves work from Azure to another cloud provider.	Not yet
M6. Projects slowed, paused or dropped for demand or returns Microsoft slowing, pausing or cancelling data centre projects, or walking away from leases and contracts for rented capacity, where the reason given is demand or returns. More of this undercuts this assessment. A change Microsoft attributes to accounting, to the timing of deliveries or to a lower cost for each unit of capacity does not count.	Weakens. A stronger factor weakens the assessment.	Microsoft lowers its capital spending outlook and cites demand or returns. Microsoft is reported to have walked away from data centre leases or contracts for rented capacity. Microsoft pauses a data centre project and cites demand.	Not yet

Assumptions

What the assessment assumes, and what would break it

MA1. Demand for Azure keeps growing faster than the capacity Microsoft can add, so new capacity earns revenue as soon as it is ready.
If this fails, the assessment fails.
Invalidated if Azure revenue growth, year over year as reported in Microsoft's quarterly earnings releases, is lower than in the previous quarter in three consecutive quarters, and on the earnings calls for all three of those quarters Microsoft does not describe its capacity as short of customer demand. If Microsoft stops reporting Azure growth, the assumption is reviewed rather than invalidated.
Source: Microsoft Investor Relations, investor metrics for the fourth quarter of fiscal year 2026, including Azure revenue growth, Microsoft Investor Relations, transcript of the earnings call for the fourth quarter of fiscal year 2026, which describes demand as exceeding available capacity. The criterion’s fingerprint was fixed on 15 October 2026 and is part of the record’s first day.
MA2. Revenue that customers have committed to Microsoft under contract keeps growing, so much of the capacity it builds is sold before it is ready.
If this fails, the assessment weakens.
Invalidated if the commercial remaining performance obligation that Microsoft reports in its quarterly earnings release, meaning the revenue that its commercial customers, AI model developers among them, have committed to under contract and Microsoft has not yet recognized, is lower than a year earlier in two consecutive quarters. If Microsoft stops reporting the figure, the assumption is reviewed rather than invalidated.
Source: Microsoft Investor Relations, earnings release for the fourth quarter of fiscal year 2026, which reports commercial remaining performance obligation up 84 per cent to US$678 billion, Microsoft Investor Relations, earnings releases, webcasts and call transcripts. The criterion’s fingerprint was fixed on 15 October 2026 and is part of the record’s first day.

Other explanations

The rival explanations

Evidence moves the assessment only when it bears differently on the assessment than on at least one of these. Evidence that fits them equally carries no weight.

MX1. Microsoft's operating profit keeps growing mainly because it charges more for software that businesses already use and holds down its other costs, while its revenue from AI does not cover the rising cost of the capacity.
No development was recorded on this assessment in the last 7 days.
Favoured by:
- Revenue per user rises after price increases while paid Copilot seats grow slowly.
- Operating profit grows while the gross margin on the Microsoft Cloud falls and Microsoft cuts costs elsewhere, such as staff.
Ruled out by:
- Paid Copilot seats and charges based on AI use grow faster than revenue per user.
- The gross margin on the Microsoft Cloud holds while AI capacity grows and list prices are unchanged.
MXC. The cost of the AI capacity Microsoft builds rises faster than the revenue that capacity brings, so that by the end of 2027 Microsoft's operating profit stops growing or Microsoft cuts back its capital spending.
This is the alternative assessment, stated word for word as the main assessment’s rival: the two cannot both be right.
No development was recorded on this assessment in the last 7 days.
Favoured by:
- Microsoft's operating profit falls from a year earlier while its capital spending keeps rising.
- Microsoft lowers its capital spending outlook and cites demand or returns.
- The gross margin on the Microsoft Cloud falls as depreciation grows faster than revenue.
Ruled out by:
- Operating profit grows year over year in every quarter through 2027 while capital spending keeps rising.

A development is recorded when the rule applies it to the assessment or sets it aside. Each count comes from the readings sealed with the changes, and a development set aside as fitting every alternative equally tells no alternative apart.

Evidence

Evidence behind changes in the last 30 days

Each item is a report or data series cited by a sealed change on this assessment. The update loop reads headlines and summaries, and no summary is shown as the evidence.

No evidence has been cited in the last 30 days.

Considered, set aside

Evidence considered and set aside

Developments the update loop read in the last 30 days and set aside because they fitted this assessment and each of its rival explanations equally. Under the rule such evidence carries no weight, so none of it moved anything.

Nothing on this assessment was set aside in the last 30 days.

Changes

Changes to this assessment

No change to this assessment yet.

The other assessments

The rest of the judgment

The question’s text on this page is as sealed on the record’s first day. Fingerprint . Fingerprint pending. Records are fingerprinted each day at 23:59 UTC.