Microsoft（MSFT）— 2026-10-28 FY27Q1 法說會逐字稿（英文原文＋中譯）
日期：2026-10-28（美國時間盤後）；FY27Q1 ＝ 2026-07 至 2026-09 逐字稿為 Microsoft earnings call 音檔 AI 轉寫，英文原文與中譯並列；專有名詞與數字以公司 press release／10-Q 為準。
校正紀錄
本檔為 AI 語音轉寫，落地前已校正下列轉寫誤植；未列出者為原文。Memo 區塊與 Transcript 區塊同步修正。 - 資本支出「$4.52 billion／45.2 億美元」→ $45.2 billion／452 億美元（量綱錯誤：同段現金支付 PP&E 394 億美元、融資租賃 74 億美元，且 FY27Q2 指引逾 500 億美元，45.2 億不合理；Memo 區塊、英文、中譯同步改） - 發言稱呼與標籤誤植：00:36:47「Bob?」→「Adam?」（Nadella 回應 Adam Wood）；00:38:56「Thanks, Adam」→「Thanks, Amy」（Adam Wood 回應 Amy Hood）；00:39:00、00:43:33 標籤 Karl Keirstead／Brad Reback → Jonathan Neilson（IR 轉場語）；00:55:53 標籤 Gabriela Borges → Operator；00:55:25「Bryn」→「Brent」（Brent Thill）；「Sakya」→「Satya」（含中譯「薩基亞」） - 財務四項交叉驗算：①營收 923 億 × 毛利率 68% ≈ 毛利 628 億；營益率 50% → 營業利益約 462 億、營業費用約 166 億（YoY +6%，與 FY27Q2 指引 185–186 億 +9% 相容）；②Agents and Infrastructure 768 億 + Devices and Consumer 155 億 ＝ 923 億 ✓，且 FY27Q2 指引 801–807 億 + 159–164 億 ＝ 960–971 億 ✓；③部門營益率 53%（768 億）與 35%（155 億）加權 ≈ 50%，與整體營益率一致 ✓（故「gross margin dollars +20%…operating margins 53%」一段屬 Agents and Infrastructure 部門，而非單指 Microsoft Cloud）；④營運現金流 580 億 − 現金支付 PP&E 394 億 ＝ 自由現金流 186 億 ✓ - 保留原文未改：安全產品名「Perception」（00:13:51、00:37:53，疑為產品名轉寫，無從判斷）；「InfraSync」「Astra」「Frontier Corps」；「Frontier and support services」部門名稱沿用轉寫（待核對）；Rolls-Royce 一段轉寫斷句錯亂但語意可辨 - 提問者（Goldman Sachs）稱 OpenAI 占本季營收 11%：為提問前提，公司未確認，保留為分析師說法
警語：此 Memo 和逐字稿為 AI 生成，可能因為音檔品質導致轉錄逐字稿有誤，或是 AI 曲解原意／幻覺導致 Memo 內容有誤，請以原始音檔為準。
Memo 亮點 • Microsoft Cloud 營收突破 630 億美元，年增率加速至 30%（根據音頻日期 20261028 推斷）。Azure 營收年增 48%，Foundry 營收年增超過三倍。營業收入增長 21%，每股盈餘為 4.89 美元，提升 18%。 [CEO - 00:03:21] [CFO - 00:19:39] [CFO - 00:23:05] • 擁有超過 50,000 個 Copilot 席位的客戶數量年增超過八倍，大眾汽車宣佈計劃將 Copilot 部署給超過 330,000 名員工，成為歐洲迄今最大規模的部署之一。 [CEO - 00:04:38] • 付費 Microsoft 365 商業席位增長 5%，ARPU 增長主要由高階席位方案帶動，包括 Microsoft 365 Copilot、E5 與 E7，後者已達數百萬付費席位。 [CFO - 00:25:20] • 營運現金流為 580 億美元，增長 29%，自由現金流為 186 億美元。透過股利與股票回購向股東返還了 102 億美元。 [CFO - 00:21:22] • 下一季度 Azure 營收成長指引為以固定匯率計算的 49-50%，預期持續的效率提升將抵銷部分供應延遲。 [CFO - 00:29:35] 總體經濟 • 本季度外匯影響大致符合指引。下一季度預計外匯對總營收、銷貨成本及營運費用成長不會有顯著影響。 [CFO - 00:20:04] [CFO - 00:28:49] 產業循環週期 • 企業採用 AI 被描述為一個長達十年的過程，目前的廣泛推動才剛開始。公司觀察到從少數客戶購買大量容量，轉向更廣泛的企業採用。 [CEO - 00:47:12] 產業供需 • Azure 的客戶需求持續超過可用供應。加速的需求促使本季度在五大洲新增 24 個資料中心。 [CFO - 00:29:52] [CEO - 00:15:06] 產業狀況 • Microsoft 觀察到 AI 在各行業的採用勢頭強勁，代理系統已在汽車、銀行及醫療等領域部署。 [CEO - 00:44:46] 獲利 • 營業收入年增 21%，以固定匯率計算增長 22%。每股盈餘為 4.89 美元，調整 OpenAI 投資影響後，年增 18%（19% 以固定匯率計算）。營業利潤率年增略升至 50%。 [CFO - 00:19:49] • Microsoft Cloud 毛利率百分比年減至 66%，主要因銷售組合轉向 Azure，部分被 Azure 產品層級的毛利率改善抵銷。 [CFO - 00:26:25] 公司展望、未來成長動能 • 下一季度整體公司營收指引為 960 億至 971 億美元，成長 18% 至 19%。營業利潤率預計較去年同期略降，持續投資於研發、產能及人才。 [CFO - 00:32:47] • Microsoft 預期在 2027 財年仍將維持自由現金流正向，並在有吸引力的機會出現時，將尋求額外資金來源。 [CFO - 00:34:31] • 資料中心基礎設施及自有應用與代理的產品改進與效率提升，增加價值並改善客戶成果，為 Microsoft 帶來持久且長期的成長動能。 [CFO - 00:34:56] 市場競爭策略 • Microsoft 允許客戶使用多個前沿模型（如 OpenAI、Anthropic）及開放權重模型，促使供應商間就每項任務競爭，避免客戶被鎖定。 [CEO - 00:50:08] 營收 • 總營收為 923 億美元，年增 19%。Microsoft Cloud 營收為 639 億美元，成長加速至 30%（固定匯率 31%）。Azure 營收增長 48%。 [CFO - 00:19:39] [CFO - 00:22:48] • 裝置與消費者營收為 155 億美元，下降 3%。扣除 TAC 後的搜尋與廣告營收增長 12%。Xbox 營收下降 2%。 [CFO - 00:27:13] 訂單 • 商業訂單減少 12%，以固定匯率計算減少 11%，主要受去年大型 Azure 承諾影響，部分被本季度大型 Azure 承諾及健康的商業銷售動作抵銷。 [CFO - 00:22:00] 產品組合 • Microsoft 的商業模式正從每席位訂閱擴展至包含消費型模式，擴大總可服務市場與平均每用戶營收。Copilot 與 GitHub 均提供訂閱與使用量計費模式。 [CEO - 00:05:20] [CEO - 00:58:16] 營收地區 • 大眾汽車宣佈計劃將 Copilot 部署給超過 330,000 名員工，成為歐洲迄今最大規模的部署之一。 [CEO - 00:04:52] 上游廠商 • Microsoft 持續以 NVIDIA 與 AMD 的最新創新技術，搭配自家 Maya 系統擴展運算資源。公司已部署首批 NVIDIA Vera Rubin GPU，對其效能感到興奮。 [CEO - 00:15:49] 下游客戶 • 來自各行業的數千名客戶已購買 E7 套件，該套件結合 Copilot、E5、Entra 與 Agent 365。知名客戶包括 Amadeus、BlackRock、Pfizer、新南威爾斯衛生部及 ZEISS。 [CEO - 00:05:09] 同業 • Microsoft 的 Copilot Fusion 使客戶能以快 45% 且成本低 75% 的方式達成前沿性能，促使 Anthropic、OpenAI 等供應商在工作負載上競爭。 [CEO - 00:49:40] 稼動率 • Microsoft 透過動態分配資源至計算、資料、Foundry 與 Copilot 的最高價值機會，優化產能利用率。Copilot 工作負載的吞吐量季增超過 50%。 [CEO - 00:14:54] 存貨庫存 • Windows OEM 營收低於預期，下降 27%（固定匯率 28%），主因去年 Windows 10 支援終止需求、OEM 與通路夥伴庫存調整及 PC 市場需求疲弱。庫存水準仍偏高。 [CFO - 00:27:47] 出貨週期與交期 • Microsoft 加快產能上線速度，最大區域新 GB300 的從碼頭到上線時間季減約 20%，這些區域佔 GPU 部署大多數。 [CEO - 00:15:15] 新產品／新技術 • Microsoft 宣佈迄今最大規模的 Copilot 更新，整合聊天、協作、自動駕駛與程式碼於一體。Copilot Fusion 帶來多模型協調，軟體開發品質提升至前沿水準，速度快 40%、成本低 77%。 [CEO - 00:03:47] [CEO - 00:06:26] • MAI Cyber 現在支援代理式 M-Dash 程式碼掃描器超過 40% 的掃描，掃描成本降低 18%。MAI Transcribe 是全球最準確且具成本效益的轉錄模型。 [CEO - 00:10:02] 產品價格 • Token 定價為 Azure 營收帶來變動性。本季度高於預期的 token 定價有利於毛利率，並優先採用能帶來良好毛利的高品質模型。 [CFO - 00:41:31] 成本毛利 • 毛利金額增長 17%，毛利率為 68%。Microsoft Cloud 毛利率百分比年減至 66%，主因銷售組合轉向 Azure，部分被產品層級毛利率改善抵銷。 [CFO - 00:19:43] [CFO - 00:26:25] 原物料／關鍵零組件 • 資本支出為 452 億美元，約四分之三用於短期資產，主要是 CPU 與 GPU，包括較高的記憶體與儲存成本。 [CFO - 00:20:40] [CFO - 00:34:08] 折舊 • 本季度融資租賃金額為 74 億美元。融資租賃的全部本金價值於租賃開始日即認列為資本支出。 [CFO - 00:21:05] 匯率 • 成長率在可用時以固定匯率提供，作為評估基礎業務表現的框架，排除外匯波動影響。 [IR - 00:01:50] 銷售／管理／研發費用 • 營運費用增加 6%，主要因持續投資於研發、運算產能及人才。 [CFO - 00:20:08] 擴廠規劃 • Microsoft 本季度在五大洲新增 24 個資料中心，以因應加速的需求。 [CEO - 00:15:06] 營業利益 • 營業收入增長 21%，以固定匯率計算增長 22%。營業利潤率年增略升至 50%。 [CFO - 00:19:49] 業外損益 • 其他收入與費用為負 2.34 億美元，略低於預期，調整 OpenAI 投資影響後。 [CFO - 00:20:31] 稅率 • 第二季有效稅率預計約為 20%。 [CFO - 00:33:44] 股利政策 • 本季度透過股利與股票回購向股東返還了 102 億美元。 [CFO - 00:21:36] 現金、現金流 • 營運現金流為 580 億美元，增長 29%。自由現金流為 186 億美元，反映較高的資本支出。 [CFO - 00:21:22] 長短期借款 • Microsoft 在籌資能力上具有顯著彈性，包括租賃及債券市場管道，雖尚未動用債券市場。公司持續維持自由現金流正向，並對每一美元資本支出的回報充滿信心。 [CFO - 01:01:02] Transcript Presentation Operator - Operator
(00:00:00)Greetings and welcome to the Microsoft Fiscal Year 2027 First Quarter Earnings Conference Call. At this time, all participants are in listen-only mode. A question-and-answer session will follow the formal presentation. If anyone requires operator assistance, please press star zero on your telephone keypad. As a reminder, this conference is being recorded. It is now my pleasure to introduce Jonathan Neilson, Vice President of Investor Relations.
(00:00:00)歡迎來到微軟 2027 財政年度第一季度財報電話會議。此時，所有參與者均處於僅聆聽模式。正式報告後將進行問答環節。如果需要操作員協助，請在電話鍵盤上按星號零。提醒您，本次會議正在錄音。現在我很高興介紹投資者關係副總裁喬納森·尼爾森。
Jonathan Neilson - Microsoft, Investor Relation
(00:00:31)Good afternoon, and thank you for joining us today. On the call with me are Satya Nadella, Chairman and Chief Executive Officer; Amy Hood, Chief Financial Officer; Alice Jolla, Chief Accounting Officer; and Brian Defoe, Deputy General Counsel and Corporate Secretary. On the Microsoft Investor Relations website, you can find our earnings press release and financial summary slide deck. These materials are intended to supplement our prepared remarks during today's call and provide reconciliation between GAAP and non-GAAP financial measures.
(00:00:31)下午好，感謝您今天的參加。與我一起參加電話會議的有：首席執行官薩提亞·納德拉；首席財務官艾米·胡德；首席會計官艾莉斯·喬拉；以及副總法律顧問兼公司秘書布萊恩·德福。在微軟投資者關係網站上，您可以找到我們的財報新聞稿和財務摘要幻燈片。這些材料旨在補充我們今天電話會議中的準備發言，並提供 GAAP 與非 GAAP 財務指標之間的調整。
Jonathan Neilson - Microsoft, Investor Relation
(00:01:06)More detailed outlook slides will be available on the Microsoft Investor Relations website when we provide outlook commentary during today's call. On this call, we will discuss certain non-GAAP items. The non-GAAP financial measures provided should not be considered substitutes for, or superior to, the financial performance measures prepared in accordance with GAAP. They are included as additional clarifying items to help investors better understand the company's first quarter performance, as well as the impact these items and events have on the financial results.
(00:01:06)當我們在今天的電話會議中提供展望評論時，將在微軟投資者關係網站上提供更詳細的展望幻燈片。在本次電話會議中，我們將討論某些非 GAAP 項目。提供的非 GAAP 財務指標不應被視為 GAAP 準則下編制的財務績效指標的替代品或優越品。它們作為額外的澄清項目，幫助投資者更好地理解公司第一季度的表現，以及這些項目和事件對財務結果的影響。
Jonathan Neilson - Microsoft, Investor Relation
(00:01:43)All growth comparisons made on the call today relate to the corresponding period of last year unless otherwise noted. We will also provide growth rates in constant currency when available, as a framework for assessing how our underlying business performed, excluding the effects of foreign currency fluctuations. Where growth rates are the same in constant currency, we will refer to the growth rate only. We will post our prepared remarks to our website immediately following the call until the complete transcript is available.
(00:01:43)今天電話會議中所做的所有增長比較均與去年同期相關，除非另有說明。我們還將在可用時提供以恆定貨幣計算的增長率，作為評估我們基礎業務表現的框架，排除外幣波動的影響。當恆定貨幣的增長率相同時，我們將僅提及增長率。我們將在電話會議結束後立即將準備好的發言發佈到我們的網站，直到完整的文字記錄可用。
Jonathan Neilson - Microsoft, Investor Relation
(00:02:16)Today's call is being webcast live and recorded. If you ask a question, it will be included in our live transmission, the transcript, and any future use of the recording. You can replay the call and view the transcript on the Microsoft Investor Relations website. During this call, we will be making forward-looking statements, which are predictions, projections, or other statements about future events. These statements are based on current expectations and assumptions that are subject to risks and uncertainties.
(00:02:16)今天的電話會議正在進行直播並錄音。如果您提出問題，它將被納入我們的直播傳輸、文字記錄以及未來的錄音使用中。您可以在微軟投資者關係網站上重播電話會議並查看文字記錄。在本次電話會議中，我們將發表前瞻性聲明，這些聲明是對未來事件的預測、預測或其他陳述。這些聲明基於當前的期望和假設，並受到風險和不確定性的影響。
Jonathan Neilson - Microsoft, Investor Relation
(00:02:50)Actual results could materially differ due to factors discussed in today's earnings press release, comments made during this conference call, and the risk factor sections of our Form 10-K, Forms 10-Q, and other reports filed with the Securities and Exchange Commission. We do not undertake any duty to update any forward-looking statement. And with that, I'll turn the call over to Satya.
(00:02:50)實際結果可能因今天的財報新聞稿中討論的因素、在本次電話會議中所作的評論以及我們的 10-K 表格、10-Q 表格和其他向證券交易委員會提交的報告中的風險因素部分而有重大差異。我們不承擔更新任何前瞻性聲明的義務。那麼，我將把電話會議交給薩提亞。
Satya Nadella - Microsoft, CEO
(00:03:16)Thank you very much, Jonathan. It was a fast start to our fiscal year. Microsoft Cloud revenue surpassed $63 billion, with growth accelerating to 30% year over year. We are focused on three priorities: designing AI to empower people, ensuring every organization has sovereignty over its own learning, and building trust with controls that ensure agent safety and security. Now, let's discuss how we are making progress on these goals as we build a new tech stack, starting with Copilot. We announced our biggest update to Copilot to date this quarter, integrating chat, co-work, autopilot, and code into one unified experience.
(00:03:16)非常感謝你，喬納森。我們的財政年度開局迅速。微軟雲收入超過 630 億美元，增長加速至 30%。我們專注於三個優先事項：設計 AI 以賦能人們，確保每個組織對自己的學習擁有主權，以及建立信任，通過控制確保代理的安全性和保障。現在，讓我們討論在建立新的技術堆棧時，我們如何在這些目標上取得進展，從 Copilot 開始。我們在本季度宣佈了對 Copilot 的最大更新，將聊天、協作、自動駕駛和代碼整合為一個統一的體驗。
Satya Nadella - Microsoft, CEO
(00:03:55)It's a single product experience across commercial and consumer segments and includes all of Office built in. We think of Copilot as the new operating system for work. Customers are increasingly turning to Copilot to accomplish real work. Automated tasks have nearly doubled since July. Daily users are up nearly 50% quarter over quarter and more than five times year over year. User satisfaction improved nearly 20% quarter over quarter, reaching another all-time high. Ninety percent of Copilot queries use auto mode, which dynamically routes tasks across models, optimizing for quality, speed, and cost.
(00:03:55)這是一個跨商業和消費者領域的單一產品體驗，並包含所有內建的 Office 功能。我們將 Copilot 視為工作的新操作系統。客戶越來越多地轉向 Copilot 來完成實際工作。自七月以來，自動化任務幾乎翻了一番。每日用戶數環比增長近 50%，同比增長超過五倍。用戶滿意度環比提高近 20%，達到歷史新高。90% 的 Copilot 查詢使用自動模式，該模式動態路由任務，優化質量、速度和成本。
Satya Nadella - Microsoft, CEO
(00:04:34)These improvements are translating into larger purchases. The number of customers with more than 50,000 Copilot seats increased over eightfold year over year, including Capgemini, HCL Tech, Hitachi Group, Truist Financial, and the UK Revenue and Customs Department. Just last week, Volkswagen announced plans to deploy Copilot to more than 330,000 employees, marking one of the largest deployments to date in Europe. We're also seeing strong momentum with E7, which combines Copilot, E5, Entra, and Agent 365.
(00:04:34)這些改進正在轉化為更大的採購。擁有超過 50,000 個 Copilot 席位的客戶數量同比增長超過八倍，包括 Capgemini、HCL Tech、日立集團、Truist Financial 和英國稅務海關總署。就在上週，大眾汽車宣佈計劃將 Copilot 部署到超過 330,000 名員工，這是迄今為止在歐洲最大的部署之一。我們還看到 E7 的強勁勢頭，該產品結合了 Copilot、E5、Entra 和 Agent 365。
Satya Nadella - Microsoft, CEO
(00:05:09)Thousands of customers across every industry have purchased the suite to date, including Amadeus, BlackRock, Pfizer, the New South Wales Health Department, and ZEISS. Our business model is expanding from per-seat subscriptions to include consumption, broadening our total addressable market and average revenue per user. The subscription delivers increasingly capable everyday intelligence to every user. Typically, IT funds this, consolidating frontier advances for everyone across the organization. As agents take on higher-value work, consumption opens up business-funded demand, with spending tied directly to the value these agents deliver.
(00:05:09)到目前為止，來自各行各業的數千名客戶已購買了該套件，包括 Amadeus、貝萊德、輝瑞、新南威爾士州衛生部和蔡司。我們的商業模式正在從按席位訂閱擴展到包括消費，擴大我們的總可尋址市場和每位用戶的平均收入。訂閱為每位用戶提供越來越強大的日常智能。通常，IT 資金支持這一點，為整個組織整合前沿進展。隨著代理承擔更高價值的工作，消費開啟了商業資金需求，支出直接與這些代理所提供的價值相關。
Satya Nadella - Microsoft, CEO
(00:05:49)The number of Copilot customers actively consuming credits through usage-based billing grew over 75% during the quarter. When it comes to software development, all major coding agents have shown up to work on GitHub. Agent-generated PRs have increased sixfold over the past six months, with agents now opening more than one-third of all PRs on the platform. We are working diligently to improve reliability as agentic activity on GitHub source intensifies. At GitHub Universe this morning, we shared plans to expand capacity and rebuild the core Git architecture to support the agentic wave.
(00:05:49)在本季度，通過基於使用的計費積分主動消費的 Copilot 客戶數量增長超過 75%。在軟體開發方面，所有主要的編碼代理都已在 GitHub 上工作。代理生成的 PR 數量在過去六個月內增加了六倍，現在代理在平台上開啟的 PR 超過所有 PR 的三分之一。隨著 GitHub 上的代理活動加強，我們正在努力提高可靠性。在今天的 GitHub Universe 上，我們分享了擴大容量和重建核心 Git 架構以支持代理浪潮的計劃。
Satya Nadella - Microsoft, CEO
(00:06:26)Also at Universe, we announced the broad availability of Copilot Fusion, which focuses on bringing multi-model orchestration to software development. It delivers frontier-level quality up to 40% faster and at up to 77% lower cost. Credit consumption from agentic features on GitHub has more than doubled since usage-based billing was implemented, and we are also helping developers secure their code and transform AI-generated code into high-quality software. GitHub Advanced Security revenue increased over 50% year over year as customers leverage its AI capabilities to scan entire repositories and fix vulnerabilities in real time.
(00:06:26)在 Universe 上，我們宣佈了 Copilot Fusion 的廣泛可用性，該功能專注於將多模型編排引入軟體開發。它提供前沿級別的質量，速度提高了 40%，成本降低了 77%。自從實施基於使用量的計費以來，GitHub 上代理功能的信用消耗已經翻了一番，我們還幫助開發者保護他們的代碼，並將 AI 生成的代碼轉化為高質量的軟體。GitHub 高級安全的收入同比增長超過 50%，因為客戶利用其 AI 能力掃描整個庫並實時修復漏洞。
Satya Nadella - Microsoft, CEO
(00:07:05)In the business application space, we are extending the vision of Copilot as the new OS for work by integrating the business logic and capabilities of Dynamics 365 directly into Copilot, enabling every organization to build their own infinite SaaS factory. This quarter, we introduced 30 new Copilot skills across Dynamics 365, allowing customers to access CRM and ERP capabilities directly within their workflow. In recruiting, more than 10 million LinkedIn members use our AI-powered job search daily, and 25,000 customers are utilizing our agentic hiring solutions.
(00:07:05)在商業應用領域，我們通過將 Dynamics 365 的商業邏輯和能力直接整合到 Copilot 中，擴展了 Copilot 作為工作新操作系統的願景，使每個組織都能建立自己的無限 SaaS 工廠。本季度，我們在 Dynamics 365 中推出了 30 種新的 Copilot 技能，允許客戶在其工作流程中直接訪問 CRM 和 ERP 功能。在招聘方面，超過 1000 萬 LinkedIn 成員每天使用我們的 AI 驅動的求職搜索，並且有 25,000 名客戶正在利用我們的代理招聘解決方案。
Satya Nadella - Microsoft, CEO
(00:07:44)Recruiters using Hiring Assistant are finding quality interview candidates 33% faster while reviewing 83% fewer profiles to identify qualified matches. In healthcare, we automated 30 million patient encounters this quarter, a 70% increase year over year, helping clinicians spend less time on administrative tasks and more time with patients. Ten providers, including Advocate Health, Mass General Brigham, Mercy, Intermountain Health, and others, have used our solutions to document over one million encounters each since the start of the calendar year.
(00:07:44)使用招聘助手的招聘人員在找到合格的面試候選人時速度提高了 33%，同時審查的簡歷數量減少了 83%。在醫療保健方面，我們本季度自動化了 3000 萬次患者接觸，同比增長 70%，幫助臨床醫生減少在行政任務上的時間，增加與患者的接觸時間。包括 Advocate Health、Mass General Brigham、Mercy、Intermountain Health 等十家提供商，自日曆年開始以來，每家都使用我們的解決方案記錄了超過一百萬次接觸。
Satya Nadella - Microsoft, CEO
(00:08:19)In science, our FDE teams are collaborating with customers to use discovery to accelerate scientific progress. A notable example is Saint-Gobain, which used discovery to accelerate the development of low-carbon construction materials. Beneath our first-party agentic experience lies the common platform. It begins with Foundry, which is becoming an increasingly important driver of our growth. Foundry helps organizations build learning loops grounded in their business context and deploy agents safely at scale.
(00:08:19)在科學領域，我們的 FDE 團隊正在與客戶合作，利用發現來加速科學進步。一個顯著的例子是聖戈班，該公司利用發現加速了低碳建築材料的開發。在我們的第一方代理體驗之下是共同平台。它始於 Foundry，這已成為我們增長的越來越重要的驅動力。Foundry 幫助組織建立基於其商業背景的學習循環，並安全地大規模部署代理。
Satya Nadella - Microsoft, CEO
(00:08:56)We now have over 120,000 Foundry customers, with revenue more than tripling year over year. Foundry's model selection and feature set, combined with its deep integration across our stack, are driving differentiation. Azure OpenAI benefited significantly last quarter with major advances in OpenAI models, and when combined with our model selection built into Foundry, customers can build solutions optimized for quality, cost, and performance. In fact, the number of customers building agentic solutions with multiple models increased fivefold year over year, and revenue from open-source models on Foundry increased tenfold.
(00:08:56)我們現在擁有超過 120,000 名 Foundry 客戶，收入同比增長超過三倍。Foundry 的模型選擇和功能集，加上其在我們堆棧中的深度整合，正在推動差異化。Azure OpenAI 在上個季度受益匪淺，OpenAI 模型取得了重大進展，當與我們內置於 Foundry 的模型選擇結合時，客戶可以構建針對質量、成本和性能優化的解決方案。實際上，使用多模型構建代理解決方案的客戶數量同比增長了五倍，來自 Foundry 的開源模型的收入增長了十倍。
Satya Nadella - Microsoft, CEO
(00:09:38)We continue to make rapid progress with our first-party models. Our goal is to advance the cost-to-outcome frontier by delivering capabilities comparable to leading models within three to six months of their release at a fraction of the training cost, allowing us to pass those cost benefits to customers. You can see this reflected in our products today. MAI Cyber now powers over 40% of the scans in our agentic M-Dash code scanner at an 18% lower cost per scan. MAI Transcribe is the most accurate and cost-effective transcription model globally, supporting Dragon Copilot's multilingual workflow.
(00:09:38)我們繼續在我們的第一方模型上取得快速進展。我們的目標是通過提供與領先模型相當的能力，在其發布後的三到六個月內以極低的訓練成本推進成本與結果的邊界，從而將這些成本效益轉嫁給客戶。你可以在我們今天的產品中看到這一點。MAI Cyber 現在為我們的代理 M-Dash 代碼掃描器提供超過 40% 的掃描，成本降低了 18%。MAI Transcribe 是全球最準確且最具成本效益的轉錄模型，支持 Dragon Copilot 的多語言工作流程。
Satya Nadella - Microsoft, CEO
(00:10:20)MAI CodeFlash had 3 million users in GitHub Copilot last month alone, helping developers maximize their AI credit usage. Microsoft Decision One is our new model for structured decision-making, outperforming both large language models and other decision models in latency and quality. Over 4,000 customers, including Atos Group, Equinor, and RBI, have already adopted it within the first month of its release. We recently completed pre-training MAI Base 2, which outperforms other open-weight models we tested with four to twelve times greater training compute efficiency.
(00:10:20)MAI CodeFlash 上個月在 GitHub Copilot 中擁有 300 萬用戶，幫助開發者最大化他們的 AI 信用使用。Microsoft Decision One 是我們的新模型，用於結構化決策，在延遲和質量上超越了大型語言模型和其他決策模型。超過 4,000 名客戶，包括 Atos Group、Equinor 和 RBI，已在其發布的第一個月內採用它。我們最近完成了 MAI Base 2 的預訓練，該模型在訓練計算效率上超越了我們測試的其他開放權重模型，效率提高了四到十二倍。
Satya Nadella - Microsoft, CEO
(00:11:01)This base model will be used in upcoming MAI models across thinking, code, and cyber domains. Beyond models, we are making the infrastructure behind our own Copilot experience available to customers through Foundry. It provides customers with a Microsoft-managed way to build agents that reason, use tools, and complete complex, long-running tasks while preserving their choice of model. Another critical layer is data. We now have 44,000 paid Fabric customers, an increase of more than 50% year over year.
(00:11:01)這個基礎模型將用於即將推出的 MAI 模型，涵蓋思考、代碼和網絡領域。除了模型，我們還通過 Foundry 向客戶提供我們自己的 Copilot 體驗背後的基礎設施。它為客戶提供了一種由微軟管理的方式來構建能夠推理、使用工具並完成複雜、長期任務的代理，同時保留他們的模型選擇。另一個關鍵層面是數據。我們現在擁有 44,000 名付費 Fabric 客戶，同比增長超過 50%。
Satya Nadella - Microsoft, CEO
(00:11:38)We are bringing Foundry and Fabric closer together so you can build your own agents directly alongside your data. Over 19,000 customers now use both Foundry and Fabric as enterprises connect and operate agents over real-time operational, analytical, and unstructured data. Customers are rapidly deploying our AI-optimized databases, including PostgreSQL, SQL DB, Hyperscale, and Cosmos DB, to provide agents with fast, secure access to the data and context they require. PostgreSQL growth accelerated for the fourth consecutive quarter, and the number of PostgreSQL customers also using Foundry increased 90% year over year.
(00:11:38)我們正在將 Foundry 和 Fabric 更加緊密地結合，以便您可以直接在數據旁邊構建自己的代理。現在有超過 19,000 名客戶同時使用 Foundry 和 Fabric，企業在實時操作、分析和非結構化數據上連接和運行代理。客戶正在快速部署我們的 AI 優化數據庫，包括 PostgreSQL、SQL DB、Hyperscale 和 Cosmos DB，以便為代理提供快速、安全的數據和上下文訪問。PostgreSQL 的增長已連續第四個季度加速，使用 Foundry 的 PostgreSQL 客戶數量同比增長了 90%。
Satya Nadella - Microsoft, CEO
(00:12:24)We also maintain one of the most important databases for any enterprise: the knowledge within Microsoft 365, encompassing people, their relationships, documents, emails, meetings, chats, projects, and other artifacts. WorkIQ makes this context available to every agentic workload. With Microsoft IQ, we are unifying WorkIQ, FabricIQ, and Foundry IQ into a single context platform for agentic systems. Nearly 60,000 customers are accessing context through Microsoft IQ, up 25% quarter over quarter. As enterprises deploy more agents, they also need to govern them at scale.
(00:12:24)我們還維護著任何企業最重要的數據庫之一：Microsoft 365 中的知識，包括人員、他們的關係、文件、電子郵件、會議、聊天、項目和其他文檔。WorkIQ 使這一上下文可用於每個代理工作負載。通過 Microsoft IQ，我們將 WorkIQ、FabricIQ 和 Foundry IQ 統一為一個單一的上下文平台用於代理系統。接近 60,000 名客戶通過 Microsoft IQ 訪問上下文，環比增長 25%。隨著企業部署更多代理，他們也需要大規模管理這些代理。
Satya Nadella - Microsoft, CEO
(00:13:13)With Agent 365, we provide a platform for AI observability, enabling customers to manage, secure, and optimize the performance and cost of their agents. Over 50 million agents are now registered, an increase of more than 30% quarter over quarter. Finally, as agents gain greater access to enterprise data, system security becomes even more critical. Our approach is to separate intelligence from authority by enforcing identity, permissions, monitoring, and containment outside the model itself.
(00:13:13)通過 Agent 365，我們提供了一個 AI 可觀察性平台，使客戶能夠管理、安全和優化其代理的性能和成本。現在註冊的代理超過 5000 萬，環比增長超過 30%。最後，隨著代理獲得更大的訪問權限企業數據，系統安全變得更加關鍵。我們的做法是通過強化身份、權限、監控和模型外的控制來將智能與權威分開。
Satya Nadella - Microsoft, CEO
(00:13:51)Last week, we expanded Perception with new capabilities to help customers secure and contain agentic insiders, including enhanced agent discovery across local and cloud environments, dynamic permissions enforcing least privileged access for each task, and continuous risk assessment. We also introduced teams of guardian agents that continuously monitor agent activity to help prevent unsafe actions before they occur. We are receiving encouraging feedback on Perception overall from preview customers, including KPMG, Nationwide Building Society, Northern Trust, and Stellantis.
(00:13:51)上週，我們擴展了 Perception，新增了幫助客戶保護和控制內部代理的能力，包括在本地和雲環境中增強的代理發現、為每個任務強制執行最小特權訪問的動態權限，以及持續的風險評估。我們還引入了持續監控代理活動的守護代理團隊，以幫助在不安全行為發生之前防止其發生。我們從預覽客戶那裡收到了對 Perception 整體的鼓舞人心的反饋，包括 KPMG、全國建築協會、北方信託和 Stellantis。
Satya Nadella - Microsoft, CEO
(00:14:33)More broadly, we are helping customers secure and govern AI use across the enterprise. Customers have used Microsoft Purview to audit over 7.6 billion agent activities to date, representing a 5x increase quarter over quarter. Underlying everything I've discussed so far is our infrastructure. We are building against a growing book of business with a fungible fleet that allows us to dynamically allocate capacity to the highest-value opportunities across compute, data, Foundry, and Copilot. Accelerating demand led us to add 24 new data centers across five continents this quarter.
(00:14:33)更廣泛地說，我們正在幫助客戶在整個企業中保護和管理 AI 的使用。客戶迄今已使用 Microsoft Purview 審計超過 76 億次代理活動，這代表著季度增長 5 倍。到目前為止，我所討論的一切都基於我們的基礎設施。我們正在建立一個不斷增長的業務書籍，擁有可靈活調配的資源，讓我們能夠動態分配容量給計算、數據、Foundry 和 Copilot 中的高價值機會。需求的加速促使我們在本季度新增了 24 個數據中心，遍佈五大洲。
Satya Nadella - Microsoft, CEO
(00:15:15)We are bringing capacity online faster, reducing dock-to-live times by about 20% quarter over quarter for new GB300s in our largest regions, which constitute the majority of GPU deployments. At the same time, we continue to extract more from the infrastructure we already have. This is achieved by optimizing across data center design, silicon systems, software, and models. For example, we increased throughput for Copilot workloads by over 50% quarter over quarter. We continue to scale our fleet with the latest innovations from NVIDIA and AMD alongside our own Maya systems.
(00:15:15)我們更快地將容量上線，將我們最大地區新 GB300 的停靠到上線時間縮短了約 20%，這些地區構成了 GPU 部署的主要部分。同時，我們繼續從我們已經擁有的基礎設施中提取更多價值。這是通過優化數據中心設計、矽系統、軟件和模型來實現的。例如，我們將 Copilot 工作負載的吞吐量提高了超過 50%。我們繼續擴大我們的資源，並引入 NVIDIA 和 AMD 的最新創新，與我們自己的 Maya 系統一起使用。
Satya Nadella - Microsoft, CEO
(00:15:57)We have deployed our first NVIDIA Vera Rubin GPUs and are very excited about it. Maya 200 is already delivering up to 1.7 times more tokens per watt across AI experiences in Teams and Copilot, and Maya 300 will drive further gains, delivering 40% more tokens per watt compared to leading GPU-based systems. Our growth extends beyond GPUs. We continue to see strong adoption of our cloud-native Cobalt CPUs, improving price performance for general-purpose cloud workloads. Overall, Azure growth was broad-based across both enterprise and small and medium business segments, spanning core cloud services and AI infrastructure.
(00:15:57)我們已經部署了首批 NVIDIA Vera Rubin GPU，對此我們感到非常興奮。Maya 200 在 Teams 和 Copilot 的 AI 體驗中，每瓦特的令牌數已經提高了 1.7 倍，而 Maya 300 將進一步提升，與領先的基於 GPU 的系統相比，每瓦特提供 40% 更多的令牌。我們的增長不僅限於 GPU。我們繼續看到我們的雲原生 Cobalt CPU 的強勁採用，改善了通用雲工作負載的價格性能。整體而言，Azure 的增長在企業和中小型企業領域都很廣泛，涵蓋了核心雲服務和 AI 基礎設施。
Satya Nadella - Microsoft, CEO
(00:16:46)Finally, let me turn to devices and consumer. In Windows, we are executing our vision to bring hybrid intelligence to every desk and home, making every PC a secure agent box. As more capable models run locally, we see an opportunity to make intelligence effectively unmetered. We are working with the broader hardware ecosystem to bring these capabilities to a new generation of PCs across AMD's Ryzen AI 4000 series, Intel Core Ultra Series 3, Qualcomm Snapdragon X2 series, and earlier this month, we announced new PCs powered by NVIDIA's RTX Spark and DGX Station for Windows.
(00:16:46)最後，讓我談談設備和消費者。在 Windows 中，我們正在執行我們的願景，將混合智能帶到每一個辦公桌和家庭，使每台 PC 成為安全的代理箱。隨著更強大的模型在本地運行，我們看到有機會使智能有效地不受限制。我們正在與更廣泛的硬件生態系統合作，將這些能力帶入新一代 PC，包括 AMD 的 Ryzen AI 4000 系列、Intel Core Ultra 系列 3、Qualcomm Snapdragon X2 系列，並且在本月初，我們宣佈了由 NVIDIA 的 RTX Spark 和 DGX Station 為 Windows 提供動力的新 PC。
Satya Nadella - Microsoft, CEO
(00:17:30)With these new devices, you can run last year's Frontier models on your laptop and last quarter's model at your desk. We are integrating Windows and Agent 365. This allows agents to operate within secure boundaries on the device. We introduced Microsoft Execution Containers, a policy-driven execution layer that governs and contains agents across Windows and other platforms, integrated with Agent 365 for governance and observability. Agents like OpenAI Codex, Anthropic Claude Code, and Meta Muse are adopting MXC, and customers such as Meta are beginning to deploy it across their enterprise environments.
(00:17:30)使用這些新設備，您可以在筆記本電腦上運行去年的 Frontier 模型，並在辦公桌上運行上季度的模型。我們正在整合 Windows 和 Agent 365。這使得代理能夠在設備的安全邊界內運行。我們推出了 Microsoft 執行容器，這是一個政策驅動的執行層，管理和控制 Windows 和其他平台上的代理，並與 Agent 365 集成以進行治理和可觀察性。像 OpenAI Codex、Anthropic Claude Code 和 Meta Muse 這樣的代理正在採用 MXC，Meta 等客戶也開始在其企業環境中部署它。
Satya Nadella - Microsoft, CEO
(00:18:15)In search and advertising, Bing and Edge once again gained market share. On LinkedIn, engagement continues to deepen across our more than 1.4 billion members, with time spent viewing content in the feed up 15% year-over-year. Finally, in gaming, we remain focused on building great games and reaching players wherever they choose to play—across console, PC, mobile, and cloud. Minecraft delivered a record quarter, reaching all-time highs in both engagement and revenue, while World of Warcraft had its largest beta in franchise history.
(00:18:15)在搜索和廣告方面，Bing 和 Edge 再次獲得了市場份額。在 LinkedIn 上，我們的 14 億多會員的參與度持續加深，查看內容的時間同比增長 15%。最後，在遊戲方面，我們仍然專注於打造優秀的遊戲，並在玩家選擇的任何地方進行接觸——無論是主機、PC、移動設備還是雲端。Minecraft 創造了創紀錄的季度，在參與度和收入上達到歷史新高，而魔獸世界則迎來了其歷史上最大的測試版。
Satya Nadella - Microsoft, CEO
(00:18:54)We enter the holiday season with one of our strongest first- and third-party lineups in years, including Modern Warfare 4, which launched last week to the best reviews for a Call of Duty game in more than a decade, and Grand Theft Auto VI, coming to Xbox next month. In closing, based on the demand we see, the strength of our products, and the efficiency we are driving across the stack, I've never been more confident in the long-term returns on our investments. With that, let me turn it over to Amy.
(00:18:54)我們以多年來最強的第一方和第三方遊戲陣容之一進入假日季節，包括上週推出的《現代戰爭 4》，該遊戲獲得了十多年來《決勝時刻》遊戲的最佳評價，以及下個月將在 Xbox 上推出的《俠盜獵車手 VI》。最後，根據我們看到的需求、產品的強勁以及我們在整個堆棧中推動的效率，我從未對我們投資的長期回報感到如此自信。有鑑於此，讓我把時間交給 Amy。
Amy Hood - Microsoft, CFO
(00:19:26)Thank you, Satya, and good afternoon, everyone. Our results exceeded expectations across revenue, operating income, and earnings per share, driven by strong demand across the Microsoft Cloud. This quarter, revenue was $92.3 billion, up 19%. Gross margin dollars increased 17%, and gross margin percentage was 68%. Operating income increased 21%, and 22% in constant currency. Earnings per share was $4.89, an increase of 18% and 19% in constant currency when adjusted for the impact of our investment in OpenAI.
(00:19:26)謝謝你，Satya，大家下午好。我們的業績在收入、營業收入和每股收益方面超出了預期，這是由於 Microsoft Cloud 的強勁需求。本季度，收入為 923 億美元，增長 19%。毛利額增加了 17%，毛利率為 68%。營業收入增長了 21%，以恆定貨幣計算增長 22%。每股收益為 4.89 美元，增長 18%，以恆定貨幣計算增長 19%，調整後考慮到我們對 OpenAI 的投資影響。
Amy Hood - Microsoft, CFO
(00:20:04)Foreign exchange was roughly in line with guidance. Operating expenses increased 6%, driven by continued investments in R&D, compute capacity, and talent. Operating margins increased slightly year-over-year to 50%, driven by strong revenue growth, platform efficiency gains, and focused investments across our portfolio, partially offset by sales mix shift to Azure. When adjusted for the impact of our investments in OpenAI, other income and expense was negative $234 million, slightly lower than expected.
(00:20:04)外匯大致符合指導預期。營業費用增長了 6%，這是由於在研發、計算能力和人才方面的持續投資。營業利潤率同比略微上升至 50%，這是由於強勁的收入增長、平台效率提升以及我們投資組閤中的集中投資，部分被銷售組合轉向 Azure 所抵消。在考慮到我們對 OpenAI 投資的影響後，其他收入和支出為負 2.34 億美元，略低於預期。
Amy Hood - Microsoft, CFO
(00:20:40)Capital expenditures were $45.2 billion, as some third-party capacity deliveries expected in Q1 were delayed. Roughly three-quarters of our capex was for short-lived assets, primarily CPUs and GPUs, including higher memory and storage costs, as we continue to invest in the infrastructure required for our customers to build and scale AI applications. The remaining spend was for long-lived assets. This quarter, finance leases amounted to $7.4 billion. As a reminder, we recognize the full principal value of finance leases in CapEx at the lease commencement date.
(00:20:40)資本支出為 452 億美元，因為預計在第一季度交付的一些第三方容量延遲了。我們的資本支出中大約四分之三用於短期資產，主要是 CPU 和 GPU，包括更高的內存和存儲成本，因為我們繼續投資於客戶構建和擴展 AI 應用所需的基礎設施。剩餘的支出用於長期資產。本季度，金融租賃總額為 74 億美元。提醒一下，我們在租賃開始日期時會在資本支出中確認金融租賃的全部本金價值。
Amy Hood - Microsoft, CFO
(00:21:17)Cash paid for PP&E totaled $39.4 billion. Cash flow from operations was $58 billion, up 29%, driven by strong cloud billings and collections. Free cash flow was $18.6 billion, reflecting higher capital expenditures. Finally, we returned $10.2 billion to shareholders through dividends and share repurchases. Now, moving to our segment results, reported under the updated segment structure announced in September. Revenue from agents and infrastructure was $76.8 billion, growing 25%, reflecting strength across the connected layers of our platform.
(00:21:17)用於不動產、廠房及設備的現金支出總計 394 億美元。營運現金流為 580 億美元，增長 29%，主要受到強勁的雲端帳單和收款推動。自由現金流為 186 億美元，反映出更高的資本支出。最後，我們通過股息和股票回購向股東返還了 102 億美元。現在，轉向我們的業務部門結果，根據九月份宣佈的更新業務結構報告。來自代理商和基礎設施的收入為 768 億美元，增長 25%，反映出我們平台各連接層的強勁表現。
Amy Hood - Microsoft, CFO
(00:22:00)Commercial bookings decreased 12% and 11% in constant currency, driven by the impact of large Azure commitments in the prior year, partially offset by healthy growth in our commercial sales motions, including a large Azure commitment this quarter. Commercial remaining performance obligation grew 73% to $677 billion, with a smaller sequential decline relative to historical seasonality due to the large Azure commitment this quarter. Our RPO has a weighted average duration of 2.3 years, and roughly 30% will be recognized as revenue in the next 12 months, up 29% year over year.
(00:22:00)商業訂單在固定匯率下減少了 12% 和 11%，受到去年大型 Azure 承諾影響，部分被我們商業銷售行動的健康增長所抵消，包括本季度的一個大型 Azure 承諾。商業剩餘履約義務增長 73%，達到 6770 億美元，與歷史季節性相比，因本季度的大型 Azure 承諾而出現較小的連續下降。我們的 RPO 加權平均期限為 2.3 年，大約 30% 將在未來 12 個月內確認為收入，同比增長 29%。
Amy Hood - Microsoft, CFO
(00:22:39)The remaining portion, recognized beyond the next 12 months, increased 98%. Now, turning to Microsoft Cloud. Revenue was $63.9 billion, with growth accelerating to 30% and 31% in constant currency, and sequential dollars added nearly doubling year over year. This reflects strong demand across Azure and Microsoft 365 commercial cloud. Azure revenue exceeded expectations, increasing 48%, despite the three peak capacity delays noted earlier, driven by Foundry and improved CPU monetization. Foundry revenue more than tripled year over year, benefiting from efficiency gains across our GPU fleet that allowed us to allocate additional capacity to customers.
(00:22:39)剩餘部分在未來 12 個月後確認的增長了 98%。現在，轉向微軟雲端。收入為 639 億美元，增長加速至 30% 和 31% 的固定匯率，並且連續美元的增加幾乎是去年同期的兩倍。這反映出 Azure 和 Microsoft 365 商業雲的強勁需求。Azure 收入超出預期，增長 48%，儘管之前提到的三次峰值容量延遲，主要受到 Foundry 和改進的 CPU 變現推動。Foundry 收入同比增長超過三倍，受益於我們 GPU 艦隊的效率提升，使我們能夠將額外容量分配給客戶。
Amy Hood - Microsoft, CFO
(00:23:30)Additionally, we saw higher revenue per token during the quarter. Across our CPU fleet, engineering efforts enabled us to identify and monetize underutilized capacity this quarter, helping us meet growing demand with existing capacity. The timing of engineering efficiencies like these, along with third-party GPU lease deliveries, our own capacity build-outs, and ARPU changes from model mix, can create quarterly variability in the year-over-year growth rate. Microsoft 365 commercial cloud revenue increased 18% and 19% in constant currency when adjusted for the prior year comparable, which benefited from one point of in-period revenue recognition.
(00:23:30)此外，我們在本季度看到每個令牌的收入增加。在我們的 CPU 艦隊中，工程努力使我們能夠識別和變現本季度未充分利用的容量，幫助我們用現有容量滿足不斷增長的需求。這類工程效率的時機，以及第三方 GPU 租賃交付、自有容量擴建和模型組合的 ARPU 變化，可能會在年增長率中造成季度變異。微軟 365 商業雲收入在調整為去年可比數據後增長 18% 和 19%，受益於一點期內收入確認。
Amy Hood - Microsoft, CFO
(00:24:17)On a reported basis, revenue growth was 17% and 18% in constant currency. Results exceeded expectations, driven by stronger GitHub Copilot adoption and usage, partially offset by weaker sales execution in our M365 E3 and standalone SKUs. We continue to evolve our sales motions to balance traditional SKU per-seat rhythms with newer premium E7 Microsoft 365 Copilot and usage-based productivity offerings. Year-over-year growth was driven by both ARPU and seats. ARPU growth was led by our premium seat-based offerings, including Microsoft 365 Copilot, E5, and E7, which has already reached millions of paid seats, with additional contribution from usage-based offerings, primarily GitHub Copilot.
(00:24:17)按報告基準，收入增長為 17% 和 18% 的固定匯率。結果超出預期，主要受到 GitHub Copilot 採用和使用增強的推動，部分被我們的 M365 E3 和獨立 SKU 的銷售執行疲弱所抵消。我們持續調整銷售策略，以平衡傳統 SKU 每席次的節奏與更新的高級 E7 Microsoft 365 Copilot和基於使用的生產力產品。同比增長受到 ARPU 和席次的推動。ARPU 增長主要來自我們的高級基於席次的產品，包括 Microsoft 365 Copilot、E5 和 E7，這些產品已經達到數百萬的付費席次，並且基於使用的產品，主要是 GitHub Copilot，也有額外貢獻。
Amy Hood - Microsoft, CFO
(00:25:11)Copilot revenue growth accelerated quarter over quarter with continued adoption across Microsoft 365, GitHub, and security. Paid Microsoft 365 commercial seats grew 5%, with some impact from the weaker sales execution noted earlier. Growth was driven by install-based expansion across all customer segments, primarily in our SMB and frontline worker offerings. Microsoft 365 consumer cloud revenue increased 17% and 16% in constant currency, down sequentially as we began to lap the benefit from the prior year's price increase.
(00:25:11)Copilot 收入增長在季度間加速，隨著 Microsoft 365、GitHub 和安全性持續採用。付費的 Microsoft 365 商業席次增長 5%，受到之前提到的銷售執行疲弱的影響。增長主要來自所有客戶群的安裝基礎擴展，特別是在我們的中小企業和前線工作者產品中。微軟 365 消費者雲收入在固定匯率下增長 17% 和 16%，隨著我們開始比較去年價格上漲的好處而出現連續下降。
Amy Hood - Microsoft, CFO
(00:25:47)Productivity and server licensing revenue decreased 3% and 2% in constant currency, primarily due to a decline in transactional purchasing as we lapped the benefit from product launches in the prior year, partially offset by a greater mix of contracts with higher in-period revenue recognition. Industry solutions cloud revenue increased 10%, with growth in Dynamics 365 and LinkedIn Talent Solutions, though booking trends continued to moderate. In frontier and support services, revenue increased 4% and 5% in constant currency, driven by support.
(00:25:47)生產力和伺服器授權收入在固定匯率下減少 3% 和 2%，主要是由於交易性購買的下降，因為我們比較了去年產品推出的好處，部分被更高的期內收入確認合約的組合所抵消。行業解決方案雲收入增長 10%，Dynamics 365 和 LinkedIn 人才解決方案增長，儘管訂單趨勢持續放緩。在前沿和支持服務中，收入在固定匯率下增長 4% 和 5%，主要受到支持的推動。
Amy Hood - Microsoft, CFO
(00:26:25)Microsoft Cloud gross margin percentage declined year over year to 66%. This was driven by sales mix shift to Azure, partially offset by product-level margin improvements, primarily in Azure. For this segment, gross margin dollars increased 20% and 21% in constant currency, while operating expenses increased 10%, driven by the R&D investments noted earlier. Operating income grew 24% and 25% in constant currency, and operating margins decreased slightly year over year to 53%, as strong revenue growth, platform efficiency gains, and focused investments across our portfolio were more than offset by sales mix shift to Azure.
(00:26:25)微軟雲的毛利率同比下降至 66%。這是由於銷售組合轉向 Azure，部分被產品層面的毛利改善所抵消，主要是在 Azure 中。在這個部門，毛利美元在固定匯率下增長 20% 和 21%，而營運費用增長 10%，主要受到之前提到的研發投資的推動。營運收入在固定匯率下增長 24% 和 25%，而營運利潤率同比略微下降至 53%，因為強勁的收入增長、平台效率提升和我們投資組合的集中投資被銷售組合轉向 Azure 所抵消。
Amy Hood - Microsoft, CFO
(00:27:10)Now, turning to devices and consumer. Revenue was $15.5 billion, declining 3%. Search and advertising revenue ex-TAC increased 12%, exceeding expectations due to higher volume and rates across LinkedIn, Edge, and Bing. Xbox revenue decreased 2%, and Xbox content and services revenue decreased 2% and 1% in constant currency, driven by declines in third-party content and Xbox Game Pass, partially offset by growth in first-party content. Windows OEM and devices revenue decreased 22%.
(00:27:10)現在，轉向設備和消費者。收入為 155 億美元，下降 3%。除 TAC 外的搜索和廣告收入增長 12%，由於 LinkedIn、Edge 和 Bing 的更高交易量和利率，超出預期。Xbox 收入減少 2%，Xbox 內容和服務收入在固定匯率下減少 2% 和 1%，主要受到第三方內容和 Xbox Game Pass 的下降影響，部分被第一方內容的增長所抵消。Windows OEM 和設備收入減少 22%。
Amy Hood - Microsoft, CFO
(00:27:47)Windows OEM revenue was below expectations, decreasing 27% and 28% in constant currency, driven by the Windows 10 end-of-support demand in the prior year, inventory drawdowns by OEM and channel partners, and weaker PC market demand. Inventory levels remain elevated. Segment gross margin dollars were relatively unchanged over the year, and operating expenses decreased by 3%, driven by Xbox, partially offset by continued investments in R&D noted earlier.
(00:27:47)Windows OEM 收入低於預期，在固定匯率下減少 27% 和 28%，主要受到去年 Windows 10 結束支持需求、OEM 和渠道合作夥伴的庫存減少以及 PC 市場需求疲弱的影響。庫存水平仍然保持在高位。各部門的毛利額在年度內相對未變，而營運支出減少了 3%，主要受到 Xbox 的推動，部分被之前提到的持續研發投資所抵消。
Amy Hood - Microsoft, CFO
(00:28:20)Operating income increased 2% and 3% in constant currency, with operating margins rising year over year to 35%, driven by lower amortization of intangible assets from the Activision acquisition, partially offset by the impact of a lower mix of Windows OEM revenue. Now, moving to the outlook for our second quarter, which, unless specifically noted otherwise, is on a U.S. dollar basis. Based on current rates, we expect foreign exchange to have no meaningful impact on total revenue, COGS, or operating expense growth, nor on revenue growth in either segment, starting with segment guidance.
(00:28:20)營運收入增長了 2%，以恆定貨幣計算增長 3%，營運利潤率同比上升至 35%，這主要是由於來自動視娛樂收購的無形資產攤銷減少，部分被 Windows OEM 收入比例下降的影響所抵消。現在，轉向我們第二季度的展望，除非特別註明，否則是以美國美元為基準。根據當前匯率，我們預計外匯對總收入、銷售成本或營運支出增長沒有實質性影響，也不會影響任何部門的收入增長，從部門指導開始。
Amy Hood - Microsoft, CFO
(00:29:02)We expect revenue from agents and infrastructure of $80.1 to $80.7 billion U.S. dollars, representing growth of 24 to 25%. In commercial bookings, we anticipate healthy growth in our commercial sales motions, though results will be impacted by significant Azure commitments signed in the prior year. Timing related to significant Azure commitments like these will cause material volatility in bookings, making quarterly growth rates less indicative of underlying sales execution.
(00:29:02)我們預計來自代理商和基礎設施的收入為 801 億至 807 億美國美元，代表增長為 24% 至 25%。在商業訂單方面，我們預計商業銷售活動將健康增長，儘管結果將受到去年簽署的重大 Azure 承諾的影響。與這些重大 Azure 承諾相關的時間安排將導致訂單的重大波動，使季度增長率對基礎銷售執行的指標性降低。
Amy Hood - Microsoft, CFO
(00:29:35)In Azure, we expect revenue growth of 49 to 50% in constant currency, with continued efficiency gains across our CPU and GPU fleet helping to offset some third-party GPU-as-a-service contracts that have been delayed to the second half. Customer demand continues to exceed available supply. Looking beyond Q2, growth will continue to reflect the timing and scale of engineering efficiencies delivered in the quarter, new capacity additions, as well as a strong prior-year comparable that included accelerating growth in the second half.
(00:29:35)在 Azure 方面，我們預計以恆定貨幣計算的收入增長為 49% 至 50%，我們的 CPU 和 GPU 設備的效率持續提升，幫助抵消一些延遲到下半年的第三方 GPU 即服務合同。客戶需求持續超過可用供應。展望第二季度以外的增長，將繼續反映本季度交付的工程效率的時間和規模，新增產能，以及去年同期的強勁可比數據，其中包括下半年的增長加速。
Amy Hood - Microsoft, CFO
(00:30:12)For Microsoft 365 commercial cloud, we expect growth of approximately 19% in constant currency, with foreign exchange expected to reduce reported growth by less than one percentage point. We anticipate continued momentum across our premium per-seat offerings and usage-based offerings, along with some improvement in execution across E3 and our standalone SKUs. Growth will be driven primarily by ARPU, with additional contribution from seats, and we continue to expect revenue growth to accelerate through the year.
(00:30:12)對於 Microsoft 365 商業雲，我們預計以恆定貨幣計算的增長約為 19%，外匯預計將使報告增長減少不到一個百分點。我們預計在我們的高端每席位產品和基於使用的產品中將持續保持動能，並且在 E3 和我們的獨立 SKU 的執行上會有一些改善。增長主要由每用戶平均收入驅動，座位數也會有額外貢獻，我們繼續預期全年收入增長將加速。
Amy Hood - Microsoft, CFO
(00:30:43)Microsoft 365 consumer cloud revenue should grow in the low teens, down sequentially as we lap the price impact noted earlier. Productivity and server licensing revenue should decline in the mid to high single digits, down sequentially due to a lower expected mix of contracts that carry higher in-period revenue recognition. Quarterly growth can vary materially based on contract mix, product launch timing, and other transactional purchasing trends. In Industry Solutions Cloud, we expect revenue growth in the mid to high single digits, with continued impact from the bookings trends noted earlier.
(00:30:43)Microsoft 365 消費者雲收入應該在低十幾個百分點增長，因為我們將面臨之前提到的價格影響的逐步比較。生產力和伺服器授權收入應該在中到高個位數範圍內下降，因為預期的合同組閤中較高的即期收入認可比例較低。季度增長可能會根據合同組合、產品推出時間和其他交易購買趨勢而有實質性變化。在行業解決方案雲中，我們預計收入增長在中到高個位數範圍內，並持續受到之前提到的訂單趨勢的影響。
Amy Hood - Microsoft, CFO
(00:31:20)In frontier and support services, we expect revenue to grow in the low to mid single digits. Microsoft Cloud gross margin percentage should be roughly 65%, down year over year, as continued improvement in product-level margins is more than offset by sales mix shift to Azure. In devices and consumer, we expect revenue to be $15.9 to $16.4 billion U.S. Search and advertising revenue at attack growth should be in the low double digits, driven by consistent trends in rate and volume, with overall search and advertising revenue expected to grow at a similar rate.
(00:31:20)在前沿和支持服務中，我們預計收入將在低到中個位數範圍內增長。Microsoft 雲的毛利率應該約為 65%，同比下降，因為產品層級毛利的持續改善被銷售組合轉向 Azure 所抵消。在設備和消費者領域，我們預計收入為 159 億至 164 億美國美元，搜索和廣告收入的增長應該在低雙位數範圍內，這是由於利率和交易量的一致趨勢驅動的，整體搜索和廣告收入預計將以相似的速度增長。
Amy Hood - Microsoft, CFO
(00:31:59)In Xbox content and services, we expect revenue growth in the mid to high single digits, up sequentially, driven by the timing of content launches across both third and first parties. With content launch patterns through the year, we expect growth to be strongest in the second quarter, with some deceleration in Q3 and further moderation in Q4. Hardware revenue in Q2 should decline year over year. Windows OEM and devices revenue should decline in the mid-30s percentage range. Growth will continue to be impacted by elevated inventory levels, PC market demand, and a prior-year comparable that benefited from Windows 10 end of support.
(00:31:59)在 Xbox 內容和服務中，我們預計收入增長在中到高個位數範圍內，並且因為第三方和第一方的內容推出時間而逐步增長。隨著全年內容推出模式的變化，我們預計增長在第二季度最強，第三季度會有所減速，第四季度進一步放緩。第二季度的硬體收入應該同比下降。Windows OEM 和設備收入應該在中 30% 的百分比範圍內下降。增長將繼續受到高庫存水平、PC 市場需求以及去年同期受益於 Windows 10 結束支持的可比數據的影響。
Amy Hood - Microsoft, CFO
(00:32:37)As in prior quarters, the range of potential outcomes remains wider than normal. We expect these trends to continue through the rest of the fiscal year. Therefore, at the total company level, revenue should be between $96 and $97.1 billion U.S. dollars, representing growth of 18 to 19 percent. We expect COGS of $32.7 to $32.9 billion U.S. dollars, representing growth of 26 to 27 percent, and operating expenses of $18.5 to $18.6 billion U.S. dollars, or growth of 9 percent, driven by continued investment in R&D capacity and talent.
(00:32:37)與之前的季度一樣，潛在結果的範圍仍然比正常情況更寬。我們預計這些趨勢將持續到本財年的其餘時間。因此，在整個公司的層面上，收入應該在 960 億至 971 億美國美元之間，代表增長為 18% 至 19%。我們預計銷售成本為 327 億至 329 億美國美元，代表增長為 26% 至 27%，營運支出為 185 億至 186 億美國美元，增長為 9%，這是由於持續對研發能力和人才的投資。
Amy Hood - Microsoft, CFO
(00:33:17)Our cost guidance across both COGS and OPEX includes an estimate for the recently announced employee voluntary exit offer. Operating margins should be down slightly year over year. Excluding any impact from our investments in OpenAI, other income and expense is expected to be roughly negative $300 million, as interest income will be more than offset by interest expense, which includes payments related to data center finance leases. We expect our Q2 effective tax rate to be approximately 20%. Next, capital expenditures.
(00:33:17)我們對銷售成本和營運支出的成本指導包括最近宣佈的員工自願離職計劃的預估。營運利潤率應該同比略微下降。排除我們對 OpenAI 的投資影響，其他收入和支出預計約為負 3 億美元，因為利息收入將被利息支出所抵消，後者包括與數據中心融資租賃相關的支付。我們預計第二季度的有效稅率約為 20%。接下來，資本支出。
Amy Hood - Microsoft, CFO
(00:33:51)In Q2, we expect capital expenditure spend to be over $50 billion. Therefore, for the calendar year, spend should be approximately $170 billion due to the three peak capacity delays noted earlier. In the second half of the fiscal year, we currently expect CapEx spend to accelerate primarily from CPUs, inclusive of higher memory and storage costs, GPUs, as well as the third-party capacity deliveries that shifted to H2. As we continue to improve the efficiencies of our infrastructure, each dollar of CapEx has the potential to support more revenue and higher returns over time.
(00:33:51)在第二季度，我們預計資本支出將超過 500 億美元。因此，對於日曆年，支出應約為 1700 億美元，這是由於之前提到的三個高峰產能延遲。在財年的下半年，我們目前預計資本支出將主要來自 CPU，包括更高的記憶體和存儲成本、GPU，以及轉移到下半年的第三方產能交付。隨著我們持續改善基礎設施的效率，每一美元的資本支出都有潛力在長期內支持更多的收入和更高的回報。
Amy Hood - Microsoft, CFO
(00:34:31)Given our strong business performance, we expect to remain free cash flow positive in FY27. To complement the cash generated by the business and preserve financial flexibility, we will access additional funding sources when it is attractive to do so. In closing, we are off to a strong start in FY27 with accelerated growth across the Microsoft Cloud. Product improvements, along with efficiency gains across our data center infrastructure and first-party apps and agents, are increasing value and improving outcomes for our customers.
(00:34:31)鑒於我們強勁的業務表現，我們預期在 FY27 仍將保持自由現金流為正。為了補充業務產生的現金並保持財務靈活性，我們將在有吸引力的情況下尋求額外的資金來源。最後，我們在 FY27 開局良好，微軟雲端業務加速增長。產品改進，加上我們數據中心基礎設施和第一方應用及代理的效率提升，正在為我們的客戶增加價值並改善結果。
Amy Hood - Microsoft, CFO
(00:35:07)Delivering that value positions us for durable, long-term growth.
(00:35:07)提供這些價值使我們能夠實現持久的長期增長。
Q&A Amy Hood - Microsoft, CFO
(00:35:12)With that, let's move to the Q&A, Jonathan.
(00:35:12)那麼，讓我們進入問答環節，Jonathan。
Jonathan Neilson - Microsoft, Investor Relation
(00:35:17)Thanks, Amy. We'll now proceed to the Q&A session. Out of respect for others on the call, we ask participants to please limit themselves to one question. Operator, could you please repeat the instructions?
(00:35:17)謝謝，Amy。現在我們將進入問答環節。為了尊重其他參與者，我們請求參與者限制自己提出一個問題。操作員，您能重複一下指示嗎？
Operator - Operator
(00:35:33)Thank you. Ladies and gentlemen, to ask a question, please press star 1 on your telephone keypad; a confirmation tone will indicate your line is in the question queue. You may press star 2 if you wish to remove your question from the queue. For participants using speaker equipment, please pick up your handset before pressing the star keys. One moment, please, as we poll for questions. Our first question comes from Adam Wood of Morgan Stanley. Please proceed.
(00:35:33)謝謝。各位女士和先生，若要提問，請在電話鍵盤上按星號 1；確認音調將表示您的線路已進入問題隊列。如果您希望將問題從隊列中移除，可以按星號 2。對於使用揚聲器設備的參與者，請在按下星號鍵之前拿起您的話筒。請稍候，我們正在進行問題調查。我們的第一個問題來自摩根士丹利的 Adam Wood。請開始。
Adam Wood - Morgan Stanley, Managing Director
(00:36:14)Thanks very much. Alongside low and expected capex and improved cloud-based volumes sequentially, can I ask, as incremental AI capacity comes online, what do you see as the more important driver of returns over the next few years: improving utilization and unit costs within Azure, or your ability to monetize more layers of each workload across Fabric, Foundry, and Copilot?
(00:36:14)非常感謝。隨著預期的資本支出較低和雲端業務量的逐步改善，我可以問一下，隨著增量 AI 能力的上線，您認為未來幾年回報的更重要驅動因素是：改善 Azure 的利用率和單位成本，還是您在 Fabric、Foundry 和 Copilot 中貨幣化每個工作負載的能力？
Satya Nadella - Microsoft, CEO
(00:36:46)Thanks for the question, Adam.
(00:36:46)謝謝你的問題，Adam。
(00:36:47)Adam? I would say all of the above. We approach this with one pool, aiming to make it as efficient and fungible as possible. That's the foundation when considering the physics of building out data centers. We apply operational rigor to bring up capacity effectively. Then, of course, we optimize how workloads run on that infrastructure. Once that is established, allocation depends on demand signals. Even in our new segment reporting, you can see this reflected in Copilot signals. The stateful database underneath is WorkIQ.
(00:36:47)Adam？我會說以上所有因素。我們以一個池子來處理這個問題，旨在使其盡可能高效和可互換。這是考慮數據中心建設物理學的基礎。我們應用操作嚴謹性來有效提升產能。然後，當然，我們優化工作負載在該基礎設施上的運行方式。一旦這一點確立，分配將取決於需求信號。即使在我們的新分部報告中，您也可以在 Copilot 信號中看到這一點的反映。底層的有狀態數據庫是 WorkIQ。
Satya Nadella - Microsoft, CEO
(00:37:53)WorkIQ is used in Foundry, where people build multi-agent systems for supply chains spanning both unstructured and structured data. It uses a Postgres database, hosted in Azure. Additionally, all agentic work is registered in Agent 365, with Perception managing its security. So, from our perspective, we view all of this as the same infrastructure. Of course, we report results across various products. But to answer your core question about where margin and leverage will come from, it will not be from any single area.
(00:37:53)WorkIQ 用於 Foundry，供人們為供應鏈構建多代理系統，涵蓋非結構化和結構化數據。它使用的是 Postgres 數據庫，託管在 Azure 上。此外，所有代理工作都在 Agent 365 中註冊，由 Perception 管理其安全性。因此，從我們的角度來看，我們將這一切視為相同的基礎設施。當然，我們在各種產品中報告結果。但要回答您關於利潤率和槓桿來源的核心問題，這不會來自任何單一領域。
Satya Nadella - Microsoft, CEO
(00:38:39)It will come from all of them, compounding as we integrate across these layers.
(00:38:39)它將來自所有這些領域，隨著我們在這些層面上的整合而複合增長。
Amy Hood - Microsoft, CFO
(00:38:45)I think that's the case. Adam, apologies if we missed the first part of your question. I believe we addressed the latter half.
(00:38:45)我認為是這樣的。Adam，如果我們錯過了您問題的第一部分，我深感抱歉。我相信我們已經解答了後半部分。
Adam Wood - Morgan Stanley, Managing Director
(00:38:56)Thanks, Amy. That was very helpful. Thank you.
(00:38:56)謝謝，Amy。這非常有幫助。謝謝。
Jonathan Neilson - Microsoft, Investor Relation
(00:39:00)All right, next question, please.
(00:39:00)好的，下一個問題，請。
Operator - Operator
(00:39:04)The next question comes from Karl Keirstead of UBS. Please proceed.
(00:39:04)下一個問題來自瑞銀的 Karl Keirstead。請開始。
Karl Keirstead - UBS, Managing Director
(00:39:11)Okay, thank you. Amy and Satya, I wonder if you could unpack what I consider a key metric from this call: the guidance for Azure growth of 49 to 50% constant currency in December. I imagine several variables contribute, such as capacity deployment, allocation to third-party customers, and OpenAI strength. Could you comment on the key drivers behind this impressive number?
(00:39:11)好的，謝謝。Amy 和 Satya，我想知道您是否可以解釋一下我認為這次通話中的一個關鍵指標：對 Azure 增長的指導為49% 到 50% 的固定匯率在十二月。我想幾個變數會影響，例如產能部署、分配給第三方客戶以及 OpenAI 的強勁表現。您能否評論一下這個令人印象深刻的數字背後的主要驅動因素？
Amy Hood - Microsoft, CFO
(00:39:44)Thank you. Thanks, Karl, for the question. I'll start by discussing Q1, which I believe will help frame expectations for Q2 and beyond. And it's not unrelated, really, to Adam's question at some level as well. In Q1, I would say the engineering teams delivered very strong work in improving efficiency at the platform layer. This is not just about the efficiency of GPUs. It encompasses the efficiency at every layer—how the app utilizes the platform, how Foundry uses the platform, and the Azure platform itself.
(00:39:44)謝謝。謝謝你，Karl，提出這個問題。我將首先討論第一季度，我相信這將有助於框定對第二季度及以後的預期。這在某種程度上與亞當的問題並不無關聯。在第一季度，我會說工程團隊在提升平台層的效率方面交出了非常出色的工作。這不僅僅是關於 GPU 的效率。它涵蓋了每一層的效率——應用如何利用平台，Foundry 如何使用平台，以及 Azure 平台本身。
Amy Hood - Microsoft, CFO
(00:40:35)Those efficiencies... They are developed steadily, week by week. You have good ideas, a lot of work, some don't pan out, but you keep at it. So, the impact in any given quarter can be somewhat unpredictable. In Q1, they did a very good job. You saw some of that reflected in our ability to allocate more resources to Foundry during the quarter. That's positive because, as I mentioned on the call, demand still exceeds supply. So, when our engineering teams deliver those efficiencies, we prioritize putting them into Foundry, as Satya highlighted, since it's critical to how customers are broadly building on what I would call the new stack.
(00:40:35)這些效率……它們是穩步發展的，逐週進步。你有好的想法，很多工作，有些不會成功，但你會持續努力。因此，在任何特定的季度，影響可能會有些不可預測。在第一季度，他們做得非常好。你可以看到這反映在我們在季度內能夠分配更多資源給 Foundry 的能力上。這是積極的，因為正如我在電話中提到的，需求仍然超過供應。因此，當我們的工程團隊交付這些效率時，我們優先將它們投入到 Foundry 中，正如薩提亞所強調的，因為這對於客戶如何廣泛構建我所稱的新堆疊至關重要。
Amy Hood - Microsoft, CFO
(00:41:31)Secondly, as you mentioned, token pricing introduces some variability. We experienced higher token pricing than expected in the quarter. High-quality models that we leverage and that yield good margins benefit us during the quarter. That was the case in Q1. And I think... Also included in that Q1 figure was the efficiency I mentioned, which applies not only to our GPU fleet but also to the combined GPU, CPU, storage, and memory workload, as Satya illustrated in response to Adam's question. It involves all components.
(00:41:31)其次，正如你提到的，代幣定價引入了一些變數。我們在季度內經歷了高於預期的代幣定價。我們利用的高質量模型，並且產生良好利潤的模型，在季度內對我們有利。第一季度就是這樣的情況。我認為……在第一季度的數字中還包括我提到的效率，這不僅適用於我們的 GPU 艦隊，還適用於結合的 GPU、CPU、存儲和內存工作負載，正如薩提亞在回答亞當的問題時所說的。它涉及所有組件。
Amy Hood - Microsoft, CFO
(00:42:18)Our ability to create additional capacity within our existing fleet gave us room to convert that capacity quite quickly in Q1. So, in Q2, we have those efficiency gains continuing, which is great. We need to find more efficiencies, which we are counting on, to meet the guidance I provided, as I have confidence in the team's work continuing from Q1 into Q2. However, we also need that work to help offset some supply we expected in Q2 that has now shifted into the second half of the year. And, as you noted, there is some uncertainty around token pricing throughout the quarter.
(00:42:18)我們在現有艦隊中創造額外容量的能力使我們在第一季度能夠迅速轉換這些容量。因此，在第二季度，我們的效率增益仍在持續，這是很好的。我們需要找到更多的效率，我們也在依賴這一點，以滿足我提供的指導，因為我對團隊在第一季度到第二季度的工作持有信心。然而，我們也需要這項工作來幫助抵消我們預期在第二季度的某些供應，這些供應現在已經轉移到下半年。而且，正如你所指出的，整個季度的代幣定價存在一些不確定性。
Amy Hood - Microsoft, CFO
(00:43:10)Those factors can change. Overall, I feel confident about the efficiency gains we observed and our ability to apply them. I feel good about providing guidance based on that. But with other variables being less predictable, I hope this explanation about the levers is helpful, Karl.
(00:43:10)這些因素可能會改變。總的來說，我對我們觀察到的效率增益以及我們應用它們的能力感到自信。我對基於此提供指導感到良好。但由於其他變數的不可預測性，我希望這對於槓桿的解釋對你有幫助，卡爾。
Karl Keirstead - UBS, Managing Director
(00:43:30)Yes, it is. Thanks, Amy.
(00:43:30)是的，這是有幫助的。謝謝，艾米。
Jonathan Neilson - Microsoft, Investor Relation
(00:43:33)Thanks, Karl. Operator, next question, please.
(00:43:33)謝謝，卡爾。操作員，請問下一個問題。
Operator - Operator
(00:43:38)The next question comes from Mark Moerdler of Bernstein Research. Please proceed.
(00:43:38)下一個問題來自 Bernstein Research 的馬克·莫爾德勒。請繼續。
Mark Moerdler - Bernstein, Senior Analyst
(00:43:45)Thank you very much for taking my question. Congratulations on the solid quarter and the strong guidance. It's impressive to see how well you are executing. I'd like to get a sense of the drivers behind the AI strength in Azure—whether it is primarily from enterprise customers versus digitally native or AI-native companies. With InfraSync starting to ramp and Microsoft becoming the go-to for enterprises, Amy, how should we think about the mix of enterprise customer workloads relative to the overall workload at this point?
(00:43:45)非常感謝你回答我的問題。恭喜你們在這個穩健的季度和強勁的指導上。看到你們執行得如此出色令人印象深刻。我想了解一下Azure 中 AI 強勁增長的驅動因素——無論是主要來自企業客戶還是數位原生或 AI 原生公司。隨著 InfraSync 開始增長，微軟成為企業的首選，艾米，我們應該如何看待企業客戶工作負載與整體工作負載的比例？
Mark Moerdler - Bernstein, Senior Analyst
(00:44:25)And Satya, with the strong growth of the Azure AI platform offering, can you provide insight into what's driving that? Is it primarily enterprise-driven? Or is it more from digitally native companies? Obviously, it's a mix of both. But any details illustrating the ramp among enterprise customers would be valuable.
(00:44:25)薩提亞，隨著 Azure AI 平台產品的強勁增長，你能提供一些驅動因素的見解嗎？主要是企業驅動的嗎？還是更多來自數位原生公司？顯然，這是兩者的混合。但任何說明企業客戶增長的細節都將是有價值的。
Satya Nadella - Microsoft, CEO
(00:44:43)Thanks for the question, Mark. Maybe I'll start, and then you can add to it. I was in Europe just last week, and it was great to see the capabilities of agentic systems, which are just— Broadly getting deployed, right? When you deliver deployments, what is their automated auto-taking, trade, coupon, and dispute management—all as a fully agentic, touchless system? Rolls-Royce deploys a system to comprehensively manage all maintenance of their... jet engines, which have 30,000 parts—a complex supply chain involving planning, execution, and inventory management—all built on Foundry agentic systems.
(00:44:43)謝謝你的問題，馬克。也許我先開始，然後你可以補充。我上週在歐洲，看到代理系統的能力非常棒，這些系統——大致上已經開始部署了，對吧？當你進行部署時，它們的自動化自動接收、交易、優惠券和爭議管理是如何運作的——這一切都是作為一個完全自主、無接觸的系統？勞斯萊斯部署了一個系統來全面管理他們的所有維護...噴氣引擎，這些引擎有 30,000 個零件——一個複雜的供應鏈，涉及規劃、執行和庫存管理——這一切都是建立在 Foundry 自主系統之上。
Satya Nadella - Microsoft, CEO
(00:45:44)RBI, an Austrian bank, showed me over 20 business processes end-to-end in... And they discussed how, whether it's for credit or other areas, they can now operate much more optimally. So the breadth, the quality, their capability, and, by the way, the multi-model nature of it. They start with a frontier model. Many of them, including Unilever, recently launched an MAI model trained using our tracers. So even the sophistication level of what I consider model optimization, workload optimization, and token optimization is impressive.
(00:45:44)RBI，一家奧地利銀行，向我展示了超過 20 個端到端的業務流程...他們討論了無論是針對信貸還是其他領域，他們現在都能以更優化的方式運作。所以這個範圍、質量、他們的能力，以及順便提一下，它的多模型特性。他們從一個前沿模型開始。包括聯合利華在內的許多公司，最近推出了一個使用我們的追蹤器訓練的 MAI 模型。因此，即使是我認為的模型優化、工作負載優化和代幣優化的複雜程度也令人印象深刻。
Satya Nadella - Microsoft, CEO
(00:46:26)Which is great to see. So the breadth is beginning. Finally, it feels like enterprise workloads. They feel to me like the database and application enterprises I grew up with are now emerging in AI. And, in fact, if anything, some of the... The shiny object distractions are waning; it's not about chasing the next big thing. It's more about, can I build a system? Can I own the weights? Can I manage my costs? Can I ensure reliability? What about security? In that sense, the enterprise breadth motion is starting.
(00:46:26)這真是令人高興的事情。所以這個範圍正在開始。最後，這感覺像是企業工作負載。在我看來，它們就像我成長過程中接觸到的數據庫和應用企業，現在正在 AI 中出現。事實上，如果有什麼不同的話，一些...令人眼花繚亂的事物分散注意力的情況正在減少；這不再是追逐下一個大事。更重要的是，我能否建立一個系統？我能否擁有權重？我能否管理我的成本？我能否確保可靠性？那麼安全性呢？在這個意義上，企業範圍的運動正在開始。
Satya Nadella - Microsoft, CEO
(00:47:12)But like everything else, enterprise adoption is not a quarter-over-quarter change. It's a decade-long process. That's what I've been waiting for, as opposed to just a few customers buying a lot of capacity and showing limited growth. That's not a sustainable business, in my view. We want breadth. We are well positioned on breadth, with large customers eager to buy from us, and we are open for business. As long as the pricing is right, we may sell to them. If not, we will be patient about building the breadth business.
(00:47:12)但和其他一切一樣，企業採用並不是一個季度對季度的變化。這是一個長達十年的過程。這就是我一直在等待的，而不是僅僅幾個客戶購買大量的容量並顯示有限的增長。在我看來，這不是一個可持續的業務。我們想要範圍。我們在範圍上處於良好的位置，有大型客戶渴望向我們購買，我們隨時歡迎業務。只要價格合適，我們可能會向他們銷售。如果不行，我們將耐心地建立範圍業務。
Jonathan Neilson - Microsoft, Investor Relation
(00:47:57)Thanks, Mark. Operator, next question, please.
(00:47:57)謝謝你，馬克。操作員，請問下一個問題。
Operator - Operator
(00:48:02)The next question comes from Gabriela Borges with Goldman Sachs. Please proceed.
(00:48:02)下一個問題來自高盛的 Gabriela Borges。請繼續。
Gabriela Borges - Goldman Sachs, Managing Director
(00:48:10)Hi, good evening. Thank you. Satya, I want to stay on the topic of breadth because... You mentioned in your prepared remarks that MAI can now be comparable to leading models within three to six months of their release. At the same time, you have OpenAI representing 11% of revenue this quarter. So maybe take a step back and share your latest thinking on how token mix and token value will evolve between frontier MAI and some of your more... sophisticated enterprise cost messaging from a proprietary AI strategy standpoint?
(00:48:10)嗨，晚上好。謝謝。Satya，我想繼續談論範圍，因為...你在準備的發言中提到，MAI 現在可以在發布後三到六個月內與領先模型相媲美。同時，你有 OpenAI 在本季度佔據 11% 的收入。所以也許退一步，分享一下你對於代幣組合和代幣價值如何在前沿 MAI 和你的一些更...複雜的企業成本傳遞從專有 AI 策略的角度會如何演變？
Gabriela Borges - Goldman Sachs, Managing Director
(00:48:46)Thank you.
(00:48:46)謝謝。
Satya Nadella - Microsoft, CEO
(00:48:47)Yeah, so I think you should consider three things in the toolbox for any workload. The first is... A new model drops with great performance, and you can measure the outcome that matters to you by using it. The outcome that matters to you by using it. People should use it. We saw that with some of the OpenAI models last quarter. They were fantastic models. We're obviously very excited about the progress OpenAI has made with Astra and what's coming next. And I thought we saw strong demand for that. Then the next step is, can I actually combine multiple frontier models...
(00:48:47)是的，所以我認為你應該考慮任何工作負載的工具箱中的三個要素。第一個是...一個新的模型推出，性能優異，你可以通過使用它來衡量對你來說重要的結果。透過使用它來衡量對你來說重要的結果。人們應該使用它。我們在上個季度看到了一些 OpenAI 模型的情況。它們是非常出色的模型。我們顯然對 OpenAI 在 Astra 上取得的進展以及接下來的發展感到非常興奮。我認為我們看到了強勁的需求。然後下一步是，我是否真的可以結合多個前沿模型...
Satya Nadella - Microsoft, CEO
(00:49:37)because these models come at very different price points. In fact, this morning, artificial analysis presented what I believe is the true benchmark, which is the coding agents using our fusion, and I think we ranked number two or three there, which is fantastic to see. This means you can now achieve frontier performance 45% faster and 75% cheaper. Who wouldn't want to take advantage of that? This applies to any coding agent user. In fact, you want Anthropic and OpenAI to compete for your workload, not just once a quarter, but for every task.
(00:49:37)因為這些模型的價格點非常不同。事實上，今天早上，人工分析展示了我認為是真正基準的內容，即使用我們融合技術的編碼代理，我認為我們在那裡排名第二或第三，這真是令人振奮的消息。這意味著您現在可以以快 45% 的速度和便宜 75% 的成本實現前沿性能。誰不想利用這一點呢？這適用於任何編碼代理用戶。事實上，您希望 Anthropic 和 OpenAI 為您的工作負載競爭，不僅僅是每季度一次，而是每個任務。
Satya Nadella - Microsoft, CEO
(00:50:19)That, to me, is the value of these routers and auto-fusion routing, and we are very excited about it. Lastly, there is another important point. Many customers are saying, look, take Unilever, for example. It's their system. Ultimately, they will take even the traces after working extensively with frontier models, and they want to train their own model. This could be open weights, or in this case, MAI models. I believe all of these options need to be available. Depending on pricing and customer needs, they will choose accordingly.
(00:50:19)對我來說，這就是這些路由器和自動融合路由的價值，我們對此感到非常興奮。最後，還有另一個重要的點。許多客戶表示，看看，以聯合利華為例。這是他們的系統。最終，他們會在與前沿模型廣泛合作後，甚至會採取一些痕跡，他們希望訓練自己的模型。這可能是開放權重，或者在這種情況下，是 MAI 模型。我相信所有這些選項都需要可用。根據定價和客戶需求，他們會相應選擇。
Satya Nadella - Microsoft, CEO
(00:51:07)There is a temporality to this. It's never a matter of starting in one place and stopping there. You might start at the frontier, move to an open-weight model, then to an MAI model, and return to the frontier for the next release after a few months. That's how dynamic these workloads are. When we think about Foundry, Fabric, or Copilot, we aim to build a model system that continuously delivers value to customers. This prevents them from being locked into any single solution.
(00:51:07)這是有時間性的。這從來不是從一個地方開始然後停在那裡的問題。您可能會從前沿開始，轉向開放權重模型，然後轉向 MAI 模型，幾個月後再回到前沿進行下一次發布。這就是這些工作負載的動態性。當我們考慮 Foundry、Fabric 或 Copilot 時，我們的目標是建立一個持續為客戶提供價值的模型系統。這防止他們被鎖定在任何單一解決方案中。
Gabriela Borges - Goldman Sachs, Managing Director
(00:51:49)Thank you for the discussion,
(00:51:49)感謝您的討論，
Jonathan Neilson - Microsoft, Investor Relation
(00:51:49)Gabriela. Operator, please proceed with the next question.
(00:51:49)Gabriela。操作員，請進行下一個問題。
Operator - Operator
(00:51:55)The next question comes from Brent Thill of Jefferies. Please go ahead.
(00:51:55)下一個問題來自 Jefferies 的 Brent Thill。請繼續。
Brent Thill - Jefferies, Managing Director
(00:52:02)Thank you, Amy. It's great to see Azure approaching 50% growth. You also have another line item nearing 50%, which is your operating margin. Many wonder how you can sustain this level of growth with 50% growth at the top and 50% at the bottom. Could you discuss the operational improvements you continue to implement to maintain such a high margin?
(00:52:02)謝謝你，Amy。很高興看到 Azure 接近 50% 的增長。您還有另一項接近 50% 的項目，即您的營運利潤率。許多人想知道您如何能夠在頂部和底部都保持 50% 的增長。您能否討論一下您持續實施的運營改進，以維持如此高的利潤率？
Amy Hood - Microsoft, CFO
(00:52:29)Thank you, Brent. Let me start by saying the best solution is revenue growth. I tend to focus there first, even though many think I start at the bottom of the P&L and work upward. The correct approach is to start at the top of the P&L and work down. Our focus has been, and this is evident in all commentary, that we change segments because we are changing how we deliver products and services to customers. We do this by delivering new and different products with significant value. These products are far more integrated than before.
(00:52:29)謝謝你，Brent。讓我先說，最好的解決方案是收入增長。我傾向於首先關注這一點，儘管許多人認為我從損益表的底部開始，然後向上工作。正確的方法是從損益表的頂部開始，然後向下工作。我們的重點一直是，這在所有評論中都很明顯，我們改變了細分市場，因為我們正在改變如何向客戶交付產品和服務。我們通過提供具有顯著價值的新產品和不同產品來做到這一點。這些產品比以前更為整合。
Amy Hood - Microsoft, CFO
(00:53:17)When you see revenue accelerate at the top line, it naturally contributes to positive margins. It also leads to positive margin dollars, which I want people to focus on. The second aspect is a real focus on efficiency. When we mention efficiency, people often think it relates only to cost. While the outcome is cost-related, the true benefit of efficiency is stability. It provides security, reliability, and delivers the highest quality unit at the lowest possible cost. This is operational excellence. When you think about delivering product excellence, which is our focus, I believe we did a very good job on product quality in the past quarter, and teams should be proud of that in many areas.
(00:53:17)當您看到頂線收入加速時，自然會促進正的利潤率。這也會導致正的利潤金額，我希望人們能夠關注這一點。第二個方面是真正關注效率。當我們提到效率時，人們常常認為這僅僅與成本有關。雖然結果與成本有關，但效率的真正好處是穩定性。它提供了安全性，可靠性，並提供最高質量的單位以最低的成本。這就是運營卓越。當您考慮提供產品卓越，這是我們的重點時，我相信我們在過去一季度的產品質量上做得非常好，團隊在許多領域應該為此感到自豪。
Amy Hood - Microsoft, CFO
(00:54:20)I believe there is a strong operational focus on driving improvements. Additionally, there is a clear emphasis on operating expenses to ensure they are directed toward building new capabilities or enhancing security and quality. We have made a deliberate decision to concentrate on these two priorities, especially given the current environment and their critical importance. When you focus on these areas, and you happen to operate in a robust market with diverse customers and a global presence, positive results follow.
(00:54:20)我相信在推動改進方面有強烈的運營重點。此外，對於運營費用有明確的重視，以確保它們用於建立新能力或增強安全性和質量。我們已經做出了明確的決定，專注於這兩個優先事項，特別是在當前環境及其關鍵重要性下。當你專注於這些領域時，而且你恰好在一個擁有多樣客戶和全球存在的強勁市場中運營，積極的結果隨之而來。
Amy Hood - Microsoft, CFO
(00:55:01)This quarter, in particular, delivering significant efficiency gains alongside revenue growth without incurring additional costs has naturally resulted in benefits, which we observed. I am confident in the fundamentals behind these numbers, which reflect a focus on the right operational levers.
(00:55:01)本季度，特別是實現顯著的效率提升與收入增長同時，沒有產生額外成本，自然帶來了我們觀察到的好處。我對這些數字背後的基本面充滿信心，這反映了對正確運營槓桿的關注。
Jonathan Neilson - Microsoft, Investor Relation
(00:55:25)Thank you, Brent. All right. Next question, please.
(00:55:25)謝謝你，Brent。好的。下一個問題，請。
Operator - Operator
(00:55:53)Our next question comes from Brad Reback.
(00:55:53)我們的下一個問題來自 Brad Reback。
Brad Reback - Stifel, Managing Director
(00:56:16)Great. Apologies for that. I did not hear my name mentioned. Regarding M365 consumption, when you shifted GitHub from a per-seat model to a consumption model, it unlocked significant revenue growth. Considering the current tools for core M365 customers on the consumption side, do you expect the revenue unlock to be rapid and require minimal sales effort, or will it demand considerable effort to activate those meters? Thank you.
(00:56:16)很好。對此我表示歉意。我沒有聽到提到我的名字。關於 M365 的消耗，當你將 GitHub 從按座位計費模式轉變為消耗模式時，這解鎖了顯著的收入增長。考慮到目前核心 M365 客戶在消耗方面的工具，你是否預期收入解鎖會迅速且需要最小的銷售努力，還是需要相當大的努力來啟動這些計量器？謝謝。
Satya Nadella - Microsoft, CEO
(00:56:54)Thanks for the question. On the coding side, I believe adoption will be faster because users are accustomed to paying... on a pay-as-you-go basis, which is common in the market. Moreover, some of the announcements we made at GitHub Universe today with Fusion offer an even more attractive approach. For example, purchasing credits and using them efficiently is a key benefit. This is the promise of GitHub Copilot Fusion. We are very optimistic about the progress in this area. Regarding Copilot itself, we aim to continuously add value to the subscription with each model update, consolidating frontier advancements for broad deployment.
(00:56:54)感謝你的提問。在編碼方面，我相信採用會更快，因為用戶習慣於按需付費...這在市場上是很常見的。此外，我們今天在 GitHub Universe 上與 Fusion 的公告提供了更具吸引力的方法。例如，購買信用並有效使用是主要好處。這就是 GitHub Copilot Fusion 的承諾。我們對這一領域的進展非常樂觀。關於 Copilot 本身，我們的目標是隨著每次模型更新不斷為訂閱增值，整合前沿的進展以便廣泛部署。
Satya Nadella - Microsoft, CEO
(00:57:57)Often, IT departments fund this as a standard issue across the organization. Simultaneously, pay-as-you-go or consumption meters are available for specific needs. Both options remain open, and we want customers to understand and optimally use these choices to their advantage. We want to avoid customers over-purchasing one option and feeling constrained by token limits, as that benefits no one. It's important they recognize the subscription's value, which continually improves. The value is dynamic, not static.
(00:57:57)通常，IT 部門將其作為組織內的標準問題進行資助。同時，按需付費或消耗計量器可用於特定需求。這兩種選擇都保持開放，我們希望客戶能理解並最佳利用這些選擇以獲得優勢。我們希望避免客戶過度購買一種選擇，並因令牌限制而感到受限，因為這對任何人都沒有好處。他們認識到訂閱的價值是重要的，這個價值不斷提升。這個價值是動態的，而不是靜態的。
Satya Nadella - Microsoft, CEO
(00:58:53)At the same time, customers can utilize the consumption meter. This level of sophistication is reflected both in our sales approach and in customers' understanding, which is coming together. We feel very positive about the progress, though there is still much to accomplish.
(00:58:53)同時，客戶可以利用消耗計量器。這種複雜程度在我們的銷售方法和客戶的理解中都有所體現，這正在逐漸融合。我們對進展感到非常積極，儘管仍有許多工作要做。
Amy Hood - Microsoft, CFO
(00:59:11)This is also why we have invested in the frontier company. Alongside changes in our sales motion, which I mentioned earlier, both elements are necessary to ensure customers fully utilize their Copilot licenses and recognize the value of consumption meters. The investment in the Frontier Development Environment organization, or Frontier Corps, as well as new initiatives within our modern— —and Microsoft 365 sales teams are designed to support this. To Satya's point, adoption outside of coding will be slower, but we continue to see encouraging signs and remain optimistic.
(00:59:11)這也是為什麼我們投資於前沿公司的原因。除了我之前提到的銷售運作變化，這兩個要素都是確保客戶充分利用其 Copilot 許可證並認識到消耗計量器價值所必需的。對於前沿開發環境組織或前沿部隊的投資，以及我們現代化的———和 Microsoft 365 銷售團隊的新舉措旨在支持這一點。對於 Satya 的觀點，編碼以外的採用會較慢，但我們仍然看到鼓舞人心的跡象，並保持樂觀。
Brad Reback - Stifel, Managing Director
(01:00:12)Thank you.
(01:00:12)謝謝。
Jonathan Neilson - Microsoft, Investor Relation
(01:00:14)Thanks, Brad. Operator, we have time for one final question.
(01:00:14)謝謝你，Brad。操作員，我們有時間進行最後一個問題。
Operator - Operator
(01:00:21)The next question is from Tal Liani with Bank of America. Please proceed.
(01:00:21)下一個問題來自美國銀行的 Tal Liani。請繼續。
Tal Liani - Bank of America, Managing Director
(01:00:30)I have a clarification regarding something you just mentioned. Did you disclose the Copilot numbers this quarter? My question is more about your capital commitment or capital strategy. You mentioned in the prepared remarks that you will consider raising capital if necessary in the future. Can you discuss your strategy regarding the balance between debt, equity, and other sources of capital? How do you plan to raise capital if the need arises? Thank you.
(01:00:30)我對你剛才提到的內容有一個澄清。你在本季度披露了 Copilot 的數字嗎？我的問題更關於你的資本承諾或資本策略。你在準備的發言中提到，如果未來有必要，你會考慮籌集資本。您能談談您在債務、股權和其他資本來源之間的平衡策略嗎？如果需要，您計劃如何籌集資金？謝謝。
Amy Hood - Microsoft, CFO
(01:01:00)Great. I'll take that question. I believe what I mentioned is that if... If the opportunity is attractive, it makes sense to pursue it. We are fortunate. We have significant flexibility in our capacity to raise financing. I've discussed this previously. One form of financing we use is leasing, which serves as a vehicle for capital. Additionally, we are fortunate to have access to the debt markets. Given our credit ratings, we have not yet tapped into that market, but it remains an option. Of course, there are other financing options available as well.
(01:01:00)很好。我來回答這個問題。我相信我提到的是，如果……如果機會具有吸引力，那麼追求它是有意義的。我們很幸運。我們在籌集資金的能力上有相當大的靈活性。我之前已經討論過這個問題。我們使用的一種融資形式是租賃，這作為資本的工具。此外，我們很幸運能夠進入債務市場。鑑於我們的信用評級，我們尚未進入該市場，但這仍然是一個選擇。當然，還有其他融資選擇可用。
Amy Hood - Microsoft, CFO
(01:01:40)Overall, we feel confident as we are free cash flow positive. We are confident in the return on every dollar of capital expenditure we invest. Therefore, we believe we have multiple options to access financing markets when conditions are favorable and attractive. To directly answer your question, we did not disclose an update on the Copilot numbers this quarter. However, we did discuss momentum, revenue, and usage trends.
(01:01:40)總的來說，我們感到自信，因為我們的自由現金流是正的。我們對每一美元資本支出的回報充滿信心。因此，我們相信在條件有利和吸引人的時候，我們有多種選擇來進入融資市場。直接回答您的問題，我們本季度沒有披露 Copilot 數字的更新。不過，我們確實討論了動能、收入和使用趨勢。
Tal Liani - Bank of America, Managing Director
(01:02:11)Thank you.
(01:02:11)謝謝。
Jonathan Neilson - Microsoft, Investor Relation
(01:02:12)Thank you, Tal. That concludes the Q&A portion of today's earnings call. Thank you for joining us today; we look forward to speaking with you all again soon. Thank you.
(01:02:12)謝謝你，Tal。這結束了今天財報電話會議的問答部分。感謝您今天的參與；我們期待不久後再次與您交談。謝謝。
Operator - Operator
(01:02:26)Thank you. This concludes today's conference. You may disconnect your lines at this time. We appreciate your participation.
(01:02:26)謝謝。這結束了今天的會議。您現在可以斷開連接。我們感謝您的參與。