甲骨文 (ORCL) 2024 Q1 法說會逐字稿

Good day, everyone, and welcome to Oracle's First Quarter 2024 Earnings Call. Today's call is being recorded, and now I would like to turn the conference over to Ken Bond. Please go ahead.

大家好，歡迎參加甲骨文 2024 年第一季財報電話會議。今天的電話會議正在錄製中，現在我想將會議交給 Ken Bond。請開始。

Thank you, Lisa, and good afternoon, everyone, and welcome to Oracle's First Quarter Fiscal Year 2024 Earnings Conference Call. A copy of the press release and financial tables, which includes a GAAP to non-GAAP reconciliation and other supplemental financial information, can be viewed and downloaded from our Investor Relations website. Additionally, a list of many customers who purchased Oracle Cloud Services or went live on Oracle Cloud recently will be available from our Investor Relations website. On the call today are Chairman and Chief Technology Officer, Larry Ellison; and Chief Executive Officer, Safra Catz.

謝謝麗莎，大家下午好！歡迎參加甲骨文2024財年第一季財報電話會議。新聞稿和財務報表（包含GAAP與非GAAP對帳表及其他補充財務資訊）的副本可在我們的投資者關係網站查看和下載。此外，我們也將在我們的投資者關係網站上提供近期購買Oracle雲端服務或上線Oracle雲端的眾多客戶名單。今天參加電話會議的嘉賓包括：董事長兼首席技術官拉里·埃里森；首席執行官薩弗拉·卡茲。

As a reminder, today's discussion will include forward-looking statements, including predictions, expectations, estimates or other information that might be considered forward-looking. Throughout today's discussion, we will present some important factors relating to our business, which may potentially affect these forward-looking statements. These forward-looking statements are also subject to risks and uncertainties that may cause actual results to differ materially from statements made today. As a result, we caution you from placing undue reliance on these forward-looking statements, and we encourage you to review our most recent reports, including our 10-K and 10-Q and any applicable amendments for a complete discussion of these factors and other risks that may affect our future results or the market price of our stock.

提醒一下，今天的討論將包含前瞻性陳述，包括預測、預期、估計或其他可能被視為前瞻性的資訊。在今天的討論中，我們將介紹一些與我們業務相關的重要因素，這些因素可能會影響這些前瞻性陳述。這些前瞻性陳述也受風險和不確定性的影響，可能導致實際結果與今天的陳述有重大差異。因此，我們提醒您不要過度依賴這些前瞻性陳述，並建議您查看我們最新的報告，包括我們的10-K和10-Q報表及其任何適用的修訂版，以全面了解這些因素以及其他可能影響我們未來業績或股票市場價格的風險。

And finally, we are not obligating ourselves to revise our results or these forward-looking statements in light of new information or future events. Before taking questions, we'll begin with a few prepared remarks. And with that, I'd like to turn the call over to Safra.

最後，我們沒有義務根據新資訊或未來事件修改我們的業績或這些前瞻性陳述。在回答問題之前，我們將先準備幾句話。現在，我想將電話交給薩夫拉。

Thanks, Ken, and good afternoon, everyone. As you know, 22 years ago today was a traumatic day for our country. Like many of you, it feels like yesterday when the country lost nearly 3,000 souls. And we at Oracle, we lost 11 employees. I remember exactly where I was when the tragedy unfolded and it is still so hard speaking about it even all these years later. Today, we honor and remember each one of the victims and heroes, and we hope that their memories are a blessing to all of us.

謝謝，肯，大家下午好。眾所周知，22年前的今天對我們國家來說是悲傷的一天。和你們中的許多人一樣，感覺就像昨天一樣，國家失去了近3000人的生命。而我們甲骨文公司，我們失去了11名員工。我清楚記得悲劇發生時我在哪裡，即使這麼多年過去了，提起這件事仍然令人難過。今天，我們緬懷每一位罹難者和英雄，並希望他們的記憶能帶給我們所有人祝福。

Now before I go to our Q1 numbers, I thought it would help to start with some of the things that are going on at Oracle that you'll be hearing about over the next couple of weeks. Next week, we have Oracle CloudWorld, which will showcase the latest innovations, including AI on OCI, the progress of Oracle Autonomous Database our multi-cloud strategy, the use of Oracle analytics throughout our portfolio to drive better decision-making and the use of generative AI to differentiate Fusion, NetSuite and our industry application.

在介紹第一季業績之前，我想先介紹一下甲骨文正在進行的一些事情，你們在接下來的幾周里會聽到這些消息。下週，我們將舉辦 Oracle CloudWorld 大會，展示最新的創新成果，包括 OCI 上的人工智慧 (AI)、Oracle 自治資料庫 (Oracle Autonomous Database) 的進展、我們的多雲策略、在整個產品組合中使用 Oracle 分析技術來推動更明智的決策，以及使用生成式人工智慧來區分 Fusion、NetSuite 和我們的產業應用程式。

Now CloudWorld is our marquee event each year where current and prospective customers take time out of their busy calendars to join us in person and share their experiences. We know that there are no better spokespeople for our products and services than our existing customers. Our innovation result directly from our development teams interacting with customers to anticipate and build the next generation of products and services. Some of the customers you will hear from next week include NVIDIA, Uber, Ascension Health, Cohere, and many, many, many more. You'll also hear from our expanding set of strategic partners that are driving the Oracle ecosystem, and this includes Amdocs, VMware, Microsoft.

如今，CloudWorld 已成為我們每年的盛事，現有客戶和潛在客戶都會從繁忙的日程中抽出時間，親自到現場，分享他們的體驗。我們深知，沒有比現有客戶更能代表我們的產品和服務的了。我們的創新直接源自於開發團隊與客戶互動，共同預測並建構下一代產品和服務。下週您將聽到來自 NVIDIA、Uber、Ascension Health、Cohere 等眾多客戶的演講。您還將聽到我們不斷成長的策略合作夥伴的聲音，這些合作夥伴正在推動 Oracle 生態系統的發展，其中包括 Amdocs、VMware 和 Microsoft。

Overall, it's remarkable the interest we are getting from the ISV community to work with Oracle. There are a lot of discussions going on and you will see more announcements shortly. From a financial standpoint, we see this customer and partner ecosystem as a leading indicator of our income statement. I've been talking with you about our revenue acceleration for some time now. In Q1, our remaining performance obligations or RPO climbed to nearly $65 billion, with the portion excluding Cerner up 11%. We have now signed several deals for OCI greater than $1 billion in total value. And the first week of Q2, we booked an additional $1.5 billion in business, which isn't even included in the Q1 numbers. Approximately 49% of total RPO is expected to be recognized as revenue over the next 12 months. My point here is that customer momentum is continuing to build. This momentum is turning into bookings, and that gives me the confidence that our annual revenue growth will continue to accelerate moving forward.

總的來說，ISV 社群對我們與 Oracle 合作的興趣非常濃厚。目前有很多討論正在進行，您很快就會看到更多公告。從財務角度來看，我們將客戶和合作夥伴生態系統視為我們損益表的領先指標。一段時間以來，我一直在與您討論我們的收入成長。在第一季度，我們的剩餘履約義務（RPO）攀升至近 650 億美元，其中不包括 Cerner 的部分增加了 11%。我們目前已與 OCI 簽署了幾筆總價值超過 10 億美元的交易。在第二季的第一周，我們又獲得了 15 億美元的業務訂單，這甚至還沒有計入第一季的數據。預計未來 12 個月，RPO 總額的約 49% 將計入收入。我的意思是，客戶成長動能持續增強。這種勢頭正在轉化為訂單，這讓我有信心，我們的年度收入成長將繼續加速。

Now to the Q1 results, which I remind you I am announcing on day 11, only because day 8, when we were ready, was a Friday and I know none of you like that. So this quarter, we saw a modest currency tailwind, but as always, I'll discuss our financials using constant currency growth rates. Clearly, Q1 was another great quarter with total revenue at the midpoint of guidance and earnings per share $0.02 above the high end of guidance and our cloud growth was 29%. Total cloud revenue, SaaS and IaaS, excluding Cerner was $4 billion, up 29%. Now including Cerner, total cloud revenue was up 29% also, at $4.6 billion and with our IaaS revenue at $1.5 billion, up 64%, and SaaS revenue of $3.1 billion, up 17%.

現在來看看第一季的業績。我之所以在第11天宣布，只是因為我們準備好的第八天是星期五，我知道你們都不喜歡這樣。本季度，我們看到了溫和的貨幣順風，但和往常一樣，我將使用固定匯率來討論我們的財務狀況。顯然，第一季又是一個出色的季度，總收入處於指導的中點，每股收益比指導的高端高出0.02美元，我們的雲端運算收入成長了29%。不包括Cerner在內的雲端運算總收入（SaaS和IaaS）為40億美元，成長了29%。現在包括Cerner在內，雲端運算總營收也成長了29%，達到46億美元，其中IaaS營收為15億美元，成長了64%，SaaS營收為31億美元，成長了17%。

Total cloud services and license support revenue for the quarter was $9.5 billion, up 12%, driven again by our strategic cloud applications, Autonomous Database and our Gen2 OCI. Application subscription revenues which includes product support were $4.5 billion, up 11%. Our strategic back-office SaaS applications now have annualized revenue of $6.9 billion and they grew 20%. Infrastructure subscription revenues, which includes license support, were $5.1 billion, up 14%. Infrastructure cloud services revenue was up 64%. Excluding legacy hosting services, Gen2 Infrastructure Cloud services revenue grew 72% with an annualized revenue of $5.6 billion.

本季雲端服務和授權支援總收入為 95 億美元，成長 12%，這再次得益於我們的策略雲端應用程式、自治資料庫和第二代 OCI。包含產品支援的應用程式訂閱收入為 45 億美元，成長 11%。我們的策略後台 SaaS 應用程式目前的年化收入為 69 億美元，成長了 20%。包含授權支援的基礎設施訂閱收入為 51 億美元，成長了 14%。基礎設施雲端服務收入成長了 64%。不包括傳統託管服務，第二代基礎設施雲端服務收入成長了 72%，年化收入為 56 億美元。

OCI consumption revenue was up 91%. Exadata cloud services revenue was up 46%, and Autonomous Database was up 42%. Database subscription services, which includes license support, were up 6%, highlighted by cloud database services, which were up 44%. Very importantly, as on-premise databases migrate to the cloud, we expect these cloud database services will be the third leg of revenue growth alongside strategic SaaS and Gen2 OCI cloud services.

OCI消費收入成長了91%。 Exadata雲端服務收入成長了46%，自治資料庫收入成長了42%。包含許可證支援的資料庫訂閱服務收入成長了6%，其中雲端資料庫服務收入成長了44%。非常重要的是，隨著本地資料庫遷移到雲端，我們預期這些雲端資料庫服務將成為繼策略性SaaS和第二代OCI雲端服務之後，營收成長的第三支柱。

Software license revenues were $0.8 billion, down 11% following an amazing Q1 last year of 19% growth, which made it a tough compare this year. So in all, total revenue for the quarter were $12.4 billion, up 8%, including Cerner, up 9% excluding Cerner.

軟體授權收入為8億美元，年減11%，而去年第一季則實現了驚人的19%的成長，這使得今年的業績難以比較。因此，本季總營收為124億美元，包括Cerner在內的營收成長8%，不包括Cerner的營收成長9%。

Shifting to margins. The gross margins for Cloud Services and License Support was 78% with IaaS gross margins improving substantially from last year. And while we've continued to build data center capacity, we've also seen our IaaS margins go higher as these new cloud regions fill up. We monitor our expenses very carefully to ensure our gross margin percentages expand as we scale up. To this point, gross profit dollars of Cloud Services and License Support grew 9% in Q1. Non-GAAP operating income was $5.1 billion, up 12% from last year. The operating margin was 41%, up from 39% last year. As we continue to benefit from economies of scale in the cloud and drive Cerner profitability to Oracle standards, we will not only continue to grow operating income but we will also grow the operating margin percentage.

轉向利潤率。雲端服務和授權支援的毛利率為78%，其中IaaS毛利率較去年大幅提升。在我們持續擴大資料中心容量的同時，隨著這些新雲端區域的容量不斷增加，IaaS的利潤率也隨之上升。我們非常謹慎地監控各項開支，以確保毛利率百分比隨著規模的擴大而不斷提升。截至目前，雲端服務和許可證支援的毛利在第一季成長了9%。非公認會計準則下的營業收入為51億美元，較去年成長12%。營業利益率為41%，高於去年的39%。隨著我們繼續受益於雲端運算的規模經濟效應，並將Cerner的獲利能力提升至甲骨文的標準，我們不僅將繼續增加營業收入，還將提高營業利潤率百分比。

The non-GAAP tax rate for the quarter was 18.8% and non-GAAP EPS was USD 1.19, up 16% in USD, up 14% in constant currency. The GAAP EPS was USD 0.86. At quarter end, we had nearly $12.1 billion in cash and marketable securities. The short-term deferred revenue balance was $11.1 billion, up 5%. Operating cash flow for the first quarter was up 9% to $7 billion, while free cash flow was up 21% to $5.7 billion, and I expect that we will see very good results in our free cash flow for the rest of the year. Over the last 4 quarters, operating cash flow was $17.7 billion, up 68%, and free cash flow was $9.5 billion, up 76%.

本季非公認會計準則稅率為18.8%，非公認會計準則每股收益為1.19美元，以美元計算成長16%，以固定匯率計算成長14%。公認會計準則每股收益為0.86美元。截至季末，我們擁有近121億美元的現金和有價證券。短期遞延收入餘額為111億美元，成長5%。第一季營運現金流成長9%，達到70億美元，自由現金流成長21%，達到57億美元。我預計，今年剩餘時間我們的自由現金流將保持良好成長動能。過去四個季度，營運現金流為177億美元，成長68%，自由現金流為95億美元，成長76%。

Capital expenditures were $8.3 billion over the last 4 quarters, and we're clearly beginning to see the cash flow benefits stemming from our cloud transformation. CapEx was $1.3 billion in Q1 as we continue to build capacity for bookings and our customers' growing needs. Given the demand we have and see in the pipeline, I expect that fiscal year 2024 CapEx will be similar to this past year's CapEx. As always, we remain careful to pace our investments appropriately and in line with booking trends, which is why our gross margins are up in our cloud business. We now have 64 cloud regions live with 44 public cloud regions around the world and another 6 being built. 12 of these public cloud regions interconnect with Microsoft Azure. We also have 9 dedicated regions live and 11 more planned, 9 security regions and 12 EU sovereign regions live with increasing demand for more of each. And of course, we have many, many Cloud at Customer implementations.

過去四個季度，資本支出為 83 億美元，我們清楚地開始看到雲端轉型帶來的現金流效益。第一季度，資本支出為 13 億美元，我們持續提升容量以滿足預訂量和客戶日益增長的需求。鑑於我們目前和未來的需求，我預計 2024 財年的資本支出將與去年持平。像往常一樣，我們始終謹慎地根據預訂趨勢調整投資節奏，這也是我們雲端業務毛利率上升的原因。我們目前已部署 64 個雲端區域，其中 44 個公有雲區域遍佈全球，另有 6 個正在建置中。其中 12 個公有雲區域與 Microsoft Azure 互聯。此外，我們還部署了 9 個專用雲區域，並計劃新增 11 個專用雲區域，9 個安全雲區域和 12 個歐盟主權雲區域，每個區域的需求都在不斷增長。當然，我們也正在實施大量的 Cloud at Customer 服務。

The cost advantages, sizing flexibility and deployment optionality of our cloud region continue to make us so compelling in the marketplace to customers. As we've said many, many times before, we're committed to returning value to our shareholders through technical innovation, strategic acquisitions, stock repurchases, prudent use of debt and a dividend. This quarter, we repurchased 1.3 million shares for a total of $150 million. In addition, we paid out dividends of $3.9 billion over the last 12 months, and the Board of Directors declared a quarterly dividend of $0.40 per share.

我們雲端區域的成本優勢、規模彈性和部署可選性，持續使我們在市場上對客戶極具吸引力。正如我們之前多次強調的那樣，我們致力於透過技術創新、策略性收購、股票回購、謹慎使用債務和股息來回報股東。本季度，我們回購了130萬股，總價值1.5億美元。此外，我們在過去12個月內派發了39億美元的股息，董事會宣布季度股息為每股0.40美元。

Now before I dive into Q2 guidance, I'd like to share some thoughts on what I see longer term. Per my earlier remarks, we have a great line of sight into the trajectory of the business, given the bookings momentum. We are extremely confident about our revenue acceleration for the year even though in any quarter, there may be small fluctuations. Because we have far more demand than we can supply, our biggest challenge is building data centers as quickly as possible. In addition, we are in an accelerated transition of Cerner to the cloud. This transition is resulting in some near-term headwinds to the Cerner growth rate as customers move from license purchases, which are recognized upfront, to cloud subscriptions which are recognized ratably. Again, excluding Cerner, I remain committed to accelerating our total revenue growth rate this fiscal year as well as maintaining our current high cloud growth rates for the year. And as you will hear at our Financial Analyst Meeting next week, we remain firmly committed to our fiscal '26 financial goals.

在深入探討第二季業績指引之前，我想先分享一些我對長期前景的看法。正如我之前所說，鑑於訂單成長勢頭強勁，我們對業務發展軌跡有著清晰的預測。儘管任何季度都可能出現小幅波動，但我們對今年的營收成長充滿信心。由於需求遠超供應能力，我們面臨的最大挑戰是如何盡快建立資料中心。此外，Cerner 正在加速向雲端轉型。隨著客戶從預先確認的授權購買模式轉向按比例確認的雲端訂閱模式，這項轉型將為 Cerner 的成長率帶來一些短期阻力。再次強調，除 Cerner 之外，我仍致力於加快本財年的總收入成長率，並維持我們目前在本財年保持較高的雲端成長率。正如您在下週的財務分析師會議上將了解的那樣，我們將繼續堅定地致力於實現 2026 財年的財務目標。

Let me now turn to my guidance for Q2, which I'll review on a non-GAAP basis. If currency exchange rates remain the same as they are now, currency should have a 2% positive effect on total revenue and a $0.03 positive effect on EPS in Q2. However, the actual currency impact may be different. Here we go. Total revenues, including Cerner, are expected to grow from 3% to 5% in constant currency and are expected to grow 5% to 7% in USD at today's rates. Total revenue, excluding Cerner, are expected to grow from 6% to 8% in constant currency and expected to grow 8% to 10% in USD.

現在，我來談談我對第二季的業績指引，我將以非公認會計準則 (Non-GAAP) 為基礎進行評估。如果匯率保持不變，匯率因素應該會對第二季的總收入產生 2% 的正向影響，並對每股盈餘產生 0.03 美元的正向影響。然而，實際的匯率影響可能會有所不同。我們來看看。包括 Cerner 在內的總營收預計以固定匯率計算成長 3% 至 5%，以美元計算成長 5% 至 7%。不包括 Cerner 在內的總營收預計以固定匯率計算成長 6% 至 8%，以美元計算成長 8% 至 10%。

Total cloud revenue, excluding Cerner, again, I give you these numbers so you can see the mainline business, is expected to grow from 27% to 29% in constant currency and is expected to grow 29% to 31% in USD. Non-GAAP EPS is expected to grow between 5% to 9% and be between $1.27 and $1.31 in constant currency. Non-GAAP EPS is expected to grow between 7% to 11% and be between USD 1.30 and USD 1.34. My EPS guidance for Q2 assumes a tax rate of 19%. However, onetime tax events could cause actual tax rates to vary. And with that, I'll turn it over to Larry for his comments.

再次強調，我提供這些數據是為了讓大家了解，不包括Cerner在內的雲端業務總收入，以固定匯率計算，預計成長27%至29%，以美元計算，預計成長29%至31%。非公認會計準則每股收益預計將成長5%至9%，以固定匯率計算，每股收益介於1.27美元至1.31美元之間。非公認會計準則每股收益預計將成長7%至11%，每股收益介於1.30美元至1.34美元之間。我對第二季每股收益的預期是基於19%的稅率。然而，一次性稅務事件可能會導致實際稅率發生變化。接下來，我將把時間交給Larry，聽聽他的意見。

Thank you, Safra. So, is generative AI the most important new computer technology ever? Maybe. We're about to find out. Self-driving cars, computer-designed antiviral drugs, voice user interfaces. Generative AI is changing the automobile industry, the pharmaceutical industry, how people communicate with their computers. Generative AI is changing everything. As of today, AI development companies have signed contracts to purchase more than $4 billion of AI training capacity in Oracle's Generation2 Cloud. That's twice as much AI training as we have booked at the end of the last Q4.

謝謝，薩弗拉。那麼，生成式人工智慧是有史以來最重要的新型電腦技術嗎？也許吧。我們即將揭曉答案。自動駕駛汽車、電腦設計的抗病毒藥物、語音使用者介面等等。生成式人工智慧正在改變汽車產業、製藥業，以及人們與電腦的溝通方式。生成式人工智慧正在改變一切。截至今天，人工智慧開發公司已簽署合同，購買甲骨文第二代雲端平台中超過 40 億美元的人工智慧訓練容量。這相當於我們在上一季末預訂的人工智慧訓練容量的兩倍。

I'm also very pleased to announce that XAI has signed a contract to do training in Oracle's Gen2 Cloud. The largest AI technology companies and the leading AI start-ups continue to expand their business with Oracle for one simple reason: Oracle's RDMA-interconnected NVIDIA superclusters train AI models at twice the speed and much less than half the cost of other clouds.

我也非常高興地宣布，XAI 已簽署合同，將在 Oracle Gen2 Cloud 上進行訓練。大型 AI 技術公司和領先的 AI 新創公司繼續與 Oracle 拓展業務，原因很簡單：Oracle 透過 RDMA 互聯的 NVIDIA 超級叢集能夠以其他雲端平台兩倍的速度訓練 AI 模型，而成本卻遠低於其他雲端平台的一半。

But growth in our AI cloud infrastructure business is not the only exciting news we have to report in Oracle. Our cloud applications business is doing quite well and it's about to get even better. In the current quarter, we expect our Cerner Health business to be awarded 2 large new contracts with a total value of over $1 billion. And I'm now also able to announce that all 9 utility companies owned by Berkshire Hathaway are in the process of replacing all their existing ERP systems and standardizing on Oracle's Fusion Cloud applications.

但AI雲端基礎設施業務的成長並非甲骨文唯一令人振奮的消息。我們的雲端應用業務表現良好，而且未來還會更好。本季度，我們預計Cerner Health業務將獲得兩份總價值超過10億美元的大型新合約。我現在還可以宣布，伯克希爾·哈撒韋旗下所有9家公用事業公司都在更換所有現有的ERP系統，並統一採用甲骨文的Fusion Cloud應用。

Let me conclude with a few words about our database business and our upcoming announcement with Microsoft later this week. We will be substantially expanding our existing multi-cloud partnership with Microsoft by making it easier for Microsoft Azure customers to buy and use the latest Oracle cloud database technology in combination with Microsoft Azure cloud services. Safra and I will discuss the details of our expanding partnership at Microsoft headquarters in Redmond on the 14th. Please tune in. And thank you. Back to you, Safra.

最後，我想簡單談談我們的資料庫業務以及本週稍後即將與微軟發布的公告。我們將大幅擴展與微軟現有的多雲合作關係，讓 Microsoft Azure 客戶能夠更輕鬆地購買和使用最新的 Oracle 雲端資料庫技術以及 Microsoft Azure 雲端服務。我和 Safra 將於 14 日在雷德蒙德的微軟總部討論我們不斷擴展的合作夥伴關係的細節。敬請關注。謝謝。 Safra，回到你這裡。

Thanks, Larry, and maybe Ken, we could start taking questions at this point.

謝謝，拉里，也許還有肯，我們現在可以開始回答問題了。

Absolutely. Before we do that, Lisa, just one quick clarification that we currently have 2 EU sovereign regions live with more to come. Lisa, please poll the audience for questions.

當然。麗莎，在此之前，我想簡單澄清一下，目前歐盟有兩個主權地區，未來還會有更多。麗莎，請對聽眾進行問卷調查，看看是否有問題。

(Operator Instructions)

（操作員指示）

We'll take our first question from Brad Zelnick with Deutsche Bank.

我們將從德意志銀行的布拉德澤爾尼克 (Brad Zelnick) 那裡得到第一個問題。

Congrats on the strong start to the year. Larry, I think it's fair to say you understand the laws of data gravity better than anyone, and you've monetized this fundamental concept as well as anyone over the years. And I recently heard someone say we're moving from a world of data gravity to one of AI gravity, and I'm not sure exactly what that means or if they even knew what that meant.

恭喜你今年開局強勁。 Larry，我認為你比任何人都更了解數據引力定律，而且多年來，你也像其他人一樣，將這個基本概念轉化為實際貨幣。最近我聽到有人說，我們正在從數據引力的世界轉向人工智慧引力的世界，我不確定這到底意味著什麼，或者他們是否真正理解這是什麼意思。

But with AI and other use cases attracting more and more data to central clouds with many vendors preaching data sharing instead of what used to be making multiple copies of things and keeping them synchronized, does AI plus cloud in any way break what we've always understood about data gravity? And what does that mean for Oracle?

但是，隨著人工智慧和其他用例將越來越多的資料吸引到中央雲端平台，許多供應商也開始倡導資料共享，而不是像過去那樣創建多個副本並保持同步。那麼，人工智慧加上雲端技術是否會以某種方式打破我們一直以來對資料引力的理解呢？這對甲骨文來說意味著什麼？

Well, you can't build any of these AI models without enormous amounts of training data. So if anything, what AI, generative AI has shown that the big issue about training one of these models is just getting -- that there's a vast amount of data ingested into your GPU supercluster. It is a huge data problem in the sense you need so much data. To train OpenAI, to train ChatGPT 3.5, they read the entire public Internet. They read all of Wikipedia. They read everything. They ingested everything.

嗯，如果沒有大量的訓練數據，你就無法建立任何這些人工智慧模型。所以，如果說有什麼區別的話，那就是生成式人工智慧表明，訓練這些模型的最大問題在於獲取——你的GPU超級叢集需要大量的資料。這是一個巨大的數據問題，因為你真的需要大量的數據。為了訓練OpenAI，為了訓練ChatGPT 3.5，他們閱讀了整個公共網路。他們閱讀了所有的維基百科。他們閱讀了所有內容。他們吸收了所有內容。

And to specialize, and you take something like ChatGPT 4.0 and you want to specialize it, you need specialized training data from electronic health records to help doctors diagnose and treat cancer, let's say. And we have Partners Imaging, for example, that is ingesting huge amounts of image data to train their AI models. We have another partner of ours in AI ingesting huge amounts of electronic health records to train their models. AI doesn't work without getting access to and ingesting enormous amounts of data.

為了實現專業化，比如說，你使用 ChatGPT 4.0 這樣的技術，並希望使其專業化，你需要從電子健康記錄中獲取專門的訓練數據，以幫助醫生診斷和治療癌症。例如，我們有 Partners Imaging，他們正在收集大量影像資料來訓練他們的人工智慧模型。我們還有另一個人工智慧合作夥伴，他們也在採集大量電子健康記錄來訓練他們的模型。如果沒有取得和採集大量數據，人工智慧就無法運作。

So in terms of a shift away from data or a change in gravity to AI, AI is utterly dependent upon vast amounts of training data. Trillion developments went into building ChatGPT 3.5. Multiple times that for ChatGPT 4.0 because you had to deal with all the image data and ingest all of that to train image recognition. So we think this is very good for our database business. And Oracle's new vector database will contain the highly specialized training data like electronic health records while keeping that data anonymized and private yet still training the specialized models that can help doctors improve their diagnostic capability and their treatment prescriptions for cancer and heart disease and all sorts of other diseases.

因此，就資料轉移或重心轉向人工智慧而言，人工智慧完全依賴大量的訓練資料。 ChatGPT 3.5 的開發投入了數兆美元的資金。 ChatGPT 4.0 的投入更是數倍，因為你必須處理所有影像資料並提取所有資料來訓練影像辨識。因此，我們認為這對我們的資料庫業務非常有利。甲骨文的新向量資料庫將包含高度專業化的訓練數據，例如電子健康記錄，同時保持這些數據的匿名性和私密性，同時仍可用於訓練專業模型，幫助醫生提高診斷能力，並製定針對癌症、心臟病和其他各種疾病的治療方案。

So we think it's a boon to our business, and we are now getting into the deepwater of the information age. Nothing has changed about that. The demands on data are getting stronger and more important.

所以我們認為這對我們的業務來說是一個福音，我們現在正進入資訊時代的深水區。這一點從未改變。對數據的需求越來越強烈，也越來越重要。

We'll take our next question from Mark Murphy with JPMorgan.

我們將回答摩根大通的馬克墨菲提出的下一個問題。

So Larry, companies are starting to understand that OCI has a very fundamentally different architecture than anything else out there in the market because of the nonblocking low-latency network design. I'm wondering if you think it's possible to actually pull farther ahead through some of your other initiatives. For instance, the Azure interconnect, it sounds like you're going to expand that having more regions, running a stronger database, providing greater isolation. Just wondering if you think there's a possibility of extending the lead.

Larry，現在許多公司開始意識到，OCI 的架構與市面上其他任何架構都截然不同，因為它採用了無阻塞低延遲網路設計。我想知道您是否認為透過其他一些舉措，OCI 能夠進一步領先。例如，Azure 互聯，聽起來您打算擴展它，擁有更多區域，運行更強大的資料庫，提供更強的隔離。您是否認為有可能進一步擴大領先優勢？

Well again, we have -- we're on our second generation of data center and our second generation of cloud. Now a lot of people noticed that we were a little bit late to the party. That's because we moved from a first generation, which we were not very happy with, to a second generation which we think solves a lot of problems the other cloud companies have not yet solved. So the nonblocking ultrafast RDMA network is not only useful for training AI models, it's useful for almost everything. It's certainly useful for building a much faster database. It's useful for -- in terms of the automation level we have in our data centers.

再說一次，我們已經進入了第二代資料中心和第二代雲端運算時代。現在很多人注意到我們有點遲到了。這是因為我們從第一代（我們對第一代不太滿意）過渡到了第二代，我們認為第二代解決了許多其他雲端運算公司尚未解決的問題。因此，無阻塞超快 RDMA 網路不僅適用於訓練 AI 模型，它幾乎適用於所有領域。它對於建立更快的資料庫當然也非常有用。就我們資料中心的自動化程度而言，它非常有用。

Our data centers are 100% automated. They configure themselves. They run themselves. We don't have a lot of labor. Now that saves us a huge amount of money, a lot of labor cost is saved. But the biggest advantage is if you don't have human beings involved, you don't have human labor, you don't have human errors. You don't have mistakes. You can ensure security. Most security problems are caused by people that make mistakes or people that engage in mischief. We don't have that in our data center. That's another huge advantage.

我們的資料中心是100%自動化的。它們可以自行配置，自行運作。我們不需要太多的勞動力。這為我們節省了大量資金，節省了大量的人力成本。但最大的優勢在於，如果沒有人工參與，就不需要人工，就不會出現人為錯誤。我們不會犯錯。這樣就能確保安全。大多數安全問題都是由人為失誤或惡意破壞造成的。而我們的資料中心沒有這種情況。這是另一個巨大的優勢。

Our data centers are -- because they're identical, the only way we could automate them was to make them all the same and they vary only by scale. There are big ones and small ones but they're identical. They all have the same hardware pieces and the same software pieces. They all have the same automation. And that automation allows us to put these data centers in very small countries. We expect to have many, many more data centers than any other cloud provider.

我們的資料中心—因為它們完全相同，我們實現自動化的唯一方法就是讓它們完全相同，它們的差異僅在於規模。我們的資料中心有大有小，但都完全相同。它們都使用相同的硬體和軟體。它們都擁有相同的自動化系統。這種自動化使我們能夠將這些資料中心設在非常小的國家。我們預計擁有的資料中心數量將比其他任何雲端服務提供者都要多得多。

But we also put those data centers at customers. Nomura Research, NRI, which resells Oracle cloud capacity in Japan, has 2 dedicated regions and are building 2 more. They run the Tokyo Stock Exchange. I don't know of any clouds that are running stock exchanges other than ours. And again, it's because of the extreme reliability and security that we get with all of the automation that's included with our data center. So we have cost advantages. We have performance advantages. We have security advantages. And that's why we're growing much faster than any of the other hyperscalers.

但我們也將這些資料中心提供給客戶。野村研究公司（NRI）在日本轉售甲骨文雲容量，擁有兩個專用區域，並且正在建造另外兩個。他們經營東京證券交易所。除了我們的雲端平台，我不知道還有其他雲端平台經營證券交易所。再說一次，這要歸功於我們資料中心內建的所有自動化功能，確保了極高的可靠性和安全性。因此，我們擁有成本優勢、性能優勢和安全優勢。這就是我們比其他任何超大規模企業成長速度都快得多的原因。

We'll take our next question from Raimo Lenschow with Barclays Capital.

我們將回答巴克萊資本的 Raimo Lenschow 提出的下一個問題。

Larry, you mentioned Berkshire and then moving over to Fusion. I just wanted to talk more in a more bigger picture on the back office systems. Like in the olden days, back office, you wouldn't touch in kind of tougher times because you had big complex projects. But you guys are still kind of growing this nicely with over 20%. Like what are you seeing there? And do you see a change, a pipeline change in customer interest of doing something there?

拉里，你提到了伯克希爾，然後又轉到了Fusion。我想更全面地談談後台系統。以前，由於專案規模龐大、複雜，在經濟困難時期，你們不會涉足後台系統。但你們現在仍然保持著超過20%的良好成長。你看到了什麼？你是否看到了客戶對後台系統的興趣改變了？

Yes. Well, the back office in the cloud is very different than the back office on-premise. And we have a big advantage that we are, by far and away, the biggest, I don't know, 95% of the cloud ERP market in terms of actual live customers using it. And we have an important partnership with JPMorgan Chase, and we'll be announcing some more partnerships in the financial community at the upcoming CloudWorld, where we automate a lot of e-commerce, B2B e-commerce right in the cloud.

是的。雲端的後台辦公室和本地的後台辦公室非常不同。我們有一個很大的優勢，那就是就實際使用雲端ERP的客戶數量而言，我們佔了目前最大的份額，大概是95%。我們與摩根大通建立了重要的合作夥伴關係，並將在即將舉行的CloudWorld大會上宣布更多金融領域的合作夥伴關係，屆時我們將在雲端實現大量電子商務、B2B電子商務的自動化。

So what is B2B e-commerce between 2 Oracle Cloud customers, 2 Oracle ERP Cloud customers? It's 1 Oracle procurement system talking to another Oracle order management system and financing the transaction through their bank. We automate that entirely in the cloud. If your bank is JPMorgan Chase, they originate the loan right along with your purchase. It's e-commerce for B2B with banking and shipping and insurance, all included enrolled together.

那麼，兩個 Oracle 雲端客戶和兩個 Oracle ERP 雲端客戶之間的 B2B 電子商務是什麼呢？它指的是一個 Oracle 採購系統與另一個 Oracle 訂單管理系統進行通信，並透過其銀行為交易提供融資。我們完全在雲端自動化了這項流程。如果您的銀行是摩根大通，他們會在您購買時立即發放貸款。這是一款包含銀行、運輸和保險服務的 B2B 電子商務，所有服務均已註冊。

No one -- we've done a great job as an industry automating e-commerce for B2C. I mean, Amazon, Walmart, others have done a brilliant job in that. We've been doing that for a long time. We have not got the equivalent in B2B commerce because B2B transactions are much more complex. In the cloud, you can get all the parties together, the shippers, the insurance company, the manufacturers, the purchasers, and we can automate that entirely within the Oracle Cloud, one ERP system talking to another, talking to their bank, talking to their insurance, doing the loan origination, getting it shipped and insured. So we make doing business much easier for our customers when they move to a modern cloud ERP system versus the on-premise ERP system that came before.

沒有人——我們在 B2C 電子商務自動化領域做得非常出色。我的意思是，亞馬遜、沃爾瑪和其他公司在這方面做得非常出色。我們在這方面已經做了很長時間。我們在 B2B 商務方面沒有同等水平，因為 B2B 交易要複雜得多。在雲端，你可以把所有參與者聚集在一起，包括托運人、保險公司、製造商、購買者，我們可以在 Oracle 雲端中完全自動化這些流程，一個 ERP 系統與另一個 ERP 系統通信，與他們的銀行通信，與他們的保險公司通信，進行貸款發放，然後發貨和投保。因此，當我們的客戶遷移到現代雲端 ERP 系統而不是以前的本地 ERP 系統時，我們可以讓他們的業務發展變得更輕鬆。

We'll take our next question from Mark Moerdler with Bernstein.

我們將回答伯恩斯坦的馬克莫德勒 (Mark Moerdler) 提出的下一個問題。

The top of mind questions I'm getting are related to AI in general as you'd expect. And more specifically as it relates to Oracle, what impact of AI will be on OCI Gen2. Two related parts to the question. The first is the profitability of AI supercomputers and whether, as some clients try to tell me, its low-calorie, empty calorie revenue, or can you maintain margins as this business grows?

正如您所料，我收到的首要問題都與人工智慧有關。更具體地說，它與甲骨文有關，人工智慧將對 OCI Gen2 產生什麼影響。這個問題有兩個相關部分。首先是人工智慧超級電腦的獲利能力，以及它是否像一些客戶試圖告訴我的那樣，只是低熱量、無用的收入，或者隨著業務的增長，你能否維持利潤率？

And the second part is about the Oracle ecosystem, and is it strong enough that as workloads transition from model training to inferencing and grounding, could AI compute create a revenue air pocket or are these kind of systems strong enough so you don't have that?

第二部分是關於 Oracle 生態系統的，它是否足夠強大，以至於隨著工作負載從模型訓練過渡到推理和基礎，AI 計算是否會造成收入空洞，或者這些系統是否足夠強大以至於不會出現這種情況？

Well, if -- you're constantly training these models. Keep in mind, you have to bring in new data if you're obviously in the health care field, in the legal field, new cases are being judged, new research is being published all the time. And for your AI models to be relevant, they have to be up to date. So it's not that you train and then do nothing but inferencing thereafter. So your training and your inferencing sit right next to each other.

嗯，如果你一直在訓練這些模型。記住，如果你在醫療保健領域，或在法律領域，你必須引入新的數據，因為新的案件在審理中，新的研究也不斷發表。為了使你的人工智慧模型具有相關性，它們必須保持最新狀態。所以，你並不是訓練完之後什麼都不做，只是進行推理。所以，你的訓練和推理是相輔相成的。

As long as we can do this stuff twice as fast as everybody else, that's on the -- by the way, not just on the training side, that's also on the inferencing side, then we're going to be half the cost or better. So we think we're going to be very, very competitive across the board, whether it's training or in inferencing. So we don't -- so we're pretty confident that we've got a cost performance advantage. Again, if you run twice as fast in the cloud, you cost half as much because you pay by the hour.

只要我們能以比其他公司快兩倍的速度完成這些工作——順便說一句，這不僅是在訓練方面，也包括推理方面，那麼我們的成本就能減半甚至更低。因此，我們認為，無論是在訓練還是推理方面，我們都能全面保持競爭力。我們非常有信心，我們擁有性價比優勢。再說一次，如果你在雲端的運行速度提高一倍，你的成本就會降低一半，因為你是按小時付費的。

So the performance advantage is really an enormous cost advantage for us. We don't see that going away anytime soon. And it applies to inferencing as well as training. Now as far as GPUs -- our GPU is a low-margin business. Not for 100% automated cloud with very, very low cost. We think in some cases, our prices for GPU training, which are very profitable by the way, for us, but are often lower -- our prices are lower than the cost of other hyperscalers doing the training.

所以效能優勢對我們來說其實是一個巨大的成本優勢。我們認為這種優勢短期內不會消失。這適用於推理和訓練。現在就GPU而言－我們的GPU業務利潤率很低。它不適用於成本極低的100%自動化雲。我們認為在某些情況下，我們的GPU訓練價格（順便說一下，這對我們來說利潤很高）通常更低——我們的價格低於其他超大規模廠商進行訓練的成本。

We'll take our next question from Keith Weiss with Morgan Stanley.

我們將回答摩根士丹利的 Keith Weiss 提出的下一個問題。

I wanted to drill in on Cerner, basically the 1-year anniversary of that acquisition. And maybe from Larry, get an update on where we are with modernizing that solution and modernizing that product. And basically, whether Cerner has kind of lived up to your expectations thus far? And then maybe for Safra, if we could dig into the expense synergy side of the equation. You guys have done a great job increasing margins on a year-on-year basis in this quarter. How much is left to go within Cerner and getting that margin profile to match the broader Oracle margin profile?

我想深入探討 Cerner，這正好是收購一周年。或許 Larry 能跟我們聊聊我們在解決方案和產品現代化上的進展。 Cerner 到目前為止是否達到了您的預期？然後，關於 Safra，我們能否深入探討費用綜效方面的問題？你們在本季實現了利潤率的同比增長，表現非常出色。 Cerner 還需要多少利潤才能達到與 Oracle 整體利潤率持平的水平？

Okay, I'll talk about the progress on taking the existing Millennium Cerner software and moving it to new Millennium, basically rewriting that software a piece at a time, by the way. It's not going to be a big rip and replace that all. There's a 2-phase process with Cerner. The first thing is to get -- to lift and shift and get the existing system hardened, which we've done and moving the customers to the cloud, which we're in the process of moving everybody to the cloud. That will give them better performance, better security and new features will then start showing up with the system.

好的，我來談談將現有的 Millennium Cerner 軟體遷移到 New Millennium 的進展，順便說一下，基本上是逐步重寫該軟體。這不會是徹底的徹底替換。 Cerner 的遷移流程分為兩個階段。第一步是提升現有系統並使其更加牢固，我們已經完成了這項工作；第二步是將客戶遷移到雲端，我們正在將所有客戶遷移到雲端。這將為他們帶來更好的性能和安全性，之後新功能也會陸續出現在系統中。

And so there's a 2-phase shift to the cloud, We're well on our way. The next is replace feature after feature after feature of the older Cerner system with a new Cerner system, new Millennium, which we are not coding in Java like we usually do. The new Cerner system is being generated, as you know, generative AI generates code. We have an application generator called APEX. And we are not writing code for the new Cerner. We are generating that code in APEX and it's going extremely well.

所以，向雲端遷移分為兩個階段，我們正在順利進行中。下一步是用新的 Cerner 系統（New Millennium）逐一取代舊版 Cerner 系統的功能。我們不像往常那樣用 Java 寫程式碼。新的 Cerner 系統正在產生程式碼，如你所知，生成式人工智慧可以產生程式碼。我們有一個名為 APEX 的應用程式產生器。我們不會為新的 Cerner 編寫程式碼，而是用 APEX 產生程式碼，而且進展非常順利。

Again, one of the great things about code generators is they don't make mistakes. Well, either they make the same mistake over and over again or they don't -- once you fixed the mistake, you fixed it everywhere. So the code -- we're using a code generator to write the new features in Cerner, and it's coming along very, very nicely. Also on the business side of things, again, we think that Cerner business is going to get stronger and stronger.

再次強調，程式碼產生器的一大優點是它們不會出錯。好吧，它們要么一遍又一遍地犯同樣的錯誤，要么就不犯——一旦你修復了錯誤，就意味著所有錯誤都修復了。所以，我們正在使用程式碼產生器來編寫 Cerner 的新功能，而且進展非常順利。同樣，在業務方面，我們認為 Cerner 的業務將會越來越強大。

Again, Safra made the point. The old Cerner business, you'd sell license so you sell a big contract and you get a big chunk of revenue in that quarter. Our new business model, as you know, is cloud. So we get a big Cerner award and we get that money now over time rather than all upfront. So that's, if you will, a bit of a revenue headwind, but the Cerner business is doing extraordinarily well.

薩弗拉再次強調了這一點。 Cerner 以前的業務是出售許可證，所以你賣出一份大合同，就能在當季獲得一大筆收入。如你所知，我們新的業務模式是雲端。所以我們獲得了一筆 Cerner 的大額合同，而且這筆錢是分期付款，而不是一次性付清。所以，如果你願意這麼說，這確實會給收入帶來一些阻力，但 Cerner 的業務表現非常出色。

And on the expense side, we still have a ways to go, but I think it will become more obvious to you next quarter, the changes we've made as they play out through the income statement more clearly. And so you'll have a better comparison Q2 to Q2, which will be a full non-deal quarter for you to look at. But you know us, we're always looking to save as much as we can and to spend as little while still really transforming Cerner into a modern system in its entirety.

在支出方面，我們還有很長的路要走，但我認為下個季度，我們所做的調整會在損益表中更加清晰地體現出來，對你們來說會更加明顯。這樣，你們就能更好地比較第二季度和第二季度，因為第二季度將是你們可以查看的完整的非交易季度。但你們知道，我們始終致力於盡可能地節省開支，同時真正地將 Cerner 轉型為一個完整的現代化系統。

Let me just reinforce what Safra just said. We love to save money. One of the things we did with our data centers is we automated them. We saved labor costs and we saved -- we have better security and better reliability because we eliminated human error. With Cerner, the rewrite of Cerner, it's not armies of programmers that are going to be rewriting this. We are generating the new Millennium software using APEX. And that's also going to save us a lot of human labor and generate higher-quality code and higher-quality user interfaces and better security all at once.

我再強調一下薩弗拉剛才說的。我們喜歡省錢。我們對資料中心所做的一件事就是實現自動化。我們節省了人力成本，而且由於消除了人為錯誤，我們的安全性和可靠性也提高了。有了 Cerner，也就是 Cerner 的重寫版，就不需要大批程式設計師來重寫了。我們正在使用 APEX 產生新的 Millennium 軟體。這也將節省大量人力，並同時產生更高品質的程式碼、更高品質的使用者介面和更強大的安全性。

We'll take our last question from John DiFucci with Guggenheim.

我們將回答古根漢美術館的 John DiFucci 提出的最後一個問題。

My question is for Safra, I think. Safra, if the organic constant currency cloud growth was in line with what you did last year and what you want to do for this year at 29%, while license, though it was a difficult comp, it was a bit weaker, at least, than the Street was expecting. And I also -- we also realized that cloud revenue for the same amount of business booked will be a lot less than the equivalent license revenue in the quarter.

我想，我的問題是問薩弗拉的。薩弗拉，如果以固定匯率計算的有機雲端業務成長率與去年和今年的29%的預期一致，那麼許可證業務雖然很難比較，但至少比華爾街預期的要弱一些。而且我還意識到，相同訂單量下的雲端業務收入將遠低於本季同等數量的授權收入。

But does this mean we're seeing a move with stronger momentum to the cloud this quarter than we have seen? And I guess just to clarify, given your guidance for the second quarter. You said you're maintaining your longer-term stuff. But I just want to clarify, are you maintaining your constant currency organic cloud guidance for the year?

但這是否意味著本季雲端運算業務的勢頭比以往更強勁？我想澄清一下，考慮到您對第二季的業績指引。您之前說過會維持長期績效指引。但我只想澄清一下，您是否會維持今年以固定匯率計算的有機雲端運算業務指引？

Yes. Okay, let me start with yes. And as I want to remind you, of course, as I want to remind you, 29% of bigger numbers is more, okay? So last year, we were smaller, and this year, we continue to plan on doing the 29%, maybe better. All of it is dependent on us getting our data centers filled up and built out as fast as possible. The level of demand we have is stunning. Stunning is the only word I can use.

是的。好的，我先說「是」。當然，我想提醒大家，29% 的數值越大越好，對吧？去年我們的規模較小，今年我們繼續計劃實現 29% 的目標，甚至可能更好。這一切都取決於我們能否盡快填滿並建立好我們的資料中心。我們的需求水準令人震驚。 「令人震驚」是我唯一能用的字眼。

And I don't want to get overly exuberant simply because we do have to continue to build out our systems, et cetera. And so yes, a very strong momentum to the cloud. And again, with the focus -- we told you, we're bringing our customers to the cloud, and that's going to have us focusing on growing that and stronger momentum there.

我不想過度興奮，因為我們確實需要繼續建立我們的系統等等。所以，是的，雲端運算的發展勢頭非常強勁。再說一次，我們重點是——我們之前說過，我們正在帶領客戶走向雲端，這將促使我們專注於發展雲端運算，並在那裡保持更強勁的發展勢頭。

Okay, great. Very clear. Yes and yes. Got it.

好的，太好了。非常清楚。是的，是的。明白了。

All right. Thank you, John, and thank you, Lisa. A telephonic replay of this conference call will be available for 24 hours on our Investor Relations website. Thank you for joining us today, and with that, I'll turn the call back to Lisa for closing.

好的。謝謝你，約翰，也謝謝你，麗莎。本次電話會議的回放將在我們投資者關係網站上提供24小時的回放。感謝大家今天的參與，現在，我將把電話會議的最後環節交還給麗莎。

That does conclude today's presentation. Thank you for your participation, and you may now disconnect.

今天的演講到此結束。感謝各位的參與，現在可以斷開連結了。