Larry Ellison's Latest Gambit to Keep Oracle on Top Is His Biggest Bet Ever

Dow Jones
10/10

By Andy Serwer

Larry Ellison is running the table.

Oracle, the giant software company he founded and controls, has been killing it, reporting a massive increase in new business from the artificial-intelligence boom. Its stock has been on fire, climbing 373% over the past three years, crushing the market (up 87%) and even its ancient foe, Microsoft (up 123%). For a brief moment, that run-up made Ellison, now chief technology officer and executive chair, the richest man on the planet. His 40.6% stake in Oracle is currently worth $343 billion.

Ellison's family life seems to have settled into a groove, too. Recently remarried, Ellison, vigorous at 81, can celebrate the success of his two grown children, Megan, 39, an accomplished film producer, and David, 42, a burgeoning media mogul. Larry Ellison has been working on deals with his son, first to buy Paramount and now a possible bid for Warner Bros. Discovery. (An ongoing offer for the U.S. operations of social-media juggernaut TikTok is by Oracle alone.)

Perhaps most gratifying to Ellison is that with AI, the world is coming around to his vision of what matters most in technology. It isn't hardware or the consumer internet or even software, it's simply data. To Ellison, harnessing, managing, and monetizing data was, is, and will always be paramount. AI, which requires by far the largest data collection and optimization endeavors in history, graphically reinforces this point.

"Larry's always had a passion for these giant processing systems running a ton of centralized data," says Zach Nelson, a former Oracle executive who became the CEO of NetSuite, a company that Ellison helped found and then bought in 2016. "That's exactly what's happening with AI. Larry's in a sweet spot now, doing what he loves the most."

Oracle has roared into the AI gold rush -- inking a $300 billion deal with OpenAI, according to The Wall Street Journal -- but with that comes huge costs and risks. On one hand, Oracle today looks perfectly positioned, and on the other, precariously perched, which isn't an unusual position for the company.

To a mere mortal CEO, this would be a recipe for disaster, but for Ellison, it's business as usual. If Ellison's ideas sound audacious, grandiose, and unfathomable, that's vintage Ellison, whose plans are sometimes processed through an Oracle dialectic, starting off with immense hype initially, then a reality check/hits-the-fan decline, and later, a working through things where the original idea is morphed and at least partially realized, often to shareholders' benefit. Oracle's AI play might be Ellison's biggest bet ever, but that's what they always say about Ellison.

"Larry's always thinking out of the box and about what's happening next, " says Tom Siebel, who joined Oracle in 1984 and later founded Siebel Systems and C3.ai. "He's enormously resilient. Think about all the changes we've had in technology. Larry has reinvented Oracle so many times. The guy is smart as hell."

"He didn't just name his company Oracle -- he is an oracle," adds Marc Benioff, co-founder and CEO of Salesforce, once a rock star salesman at Oracle.

Recall that in the late 1990s tech boom, Oracle was one of Silicon Valley's "it" companies, along with Intel, Hewlett-Packard (now Hewlett Packard Enterprise and HP Inc.), Cisco Systems, and Sun Microsystems. How do they stack up today? Oracle bought Sun in 2010, and as for the others, Oracle's market cap of $823 billion is more than 1.5 times those companies' combined.

Like his tech billionaire brethren -- Bill Gates, Elon Musk, Jeff Bezos, and Steve Jobs (the first his nemesis, the last his best friend) -- Ellison is a complicated guy. Oracle's ups and downs reflect his unparalleled vision and singular skill set, as well as his braggadocio and blind spots. Oracle's history, then, has been about swinging the bat, sometimes crushing the ball, sometimes missing, but in the latter cases, surviving to play another day.

Ellison's back story bends toward American gothic. Born to an unwed mother who gave him to relatives to raise as a baby, Ellison -- who declined to be interviewed for this article -- was estranged from his biological parents. He has said his family, immigrants from Europe, took the name "Ellison" after Ellis Island. He grew up working-class in an apartment building in Chicago. His adoptive father, Louis, was an accountant who was forever telling Larry that he was a good-for-nothing.

No wonder when Steve Jobs was asked what motivated Ellison, he chuckled and simply said, "Rosebud." And perhaps it's no wonder that Ellison now lives large. "He has a fascinating life away from business," says Dave Roux, a former Oracle executive, co-founder of Silver Lake, and now executive chairman of BayPine. "He plays classical guitar. He's a very good tennis player. He won the America's Cup. He has a huge Japanese art collection." Every spring, Ellison throws a cherry blossom festival party for his friends at his Japanese-style feudal estate in Woodside, Calif.

Ellison's real turn-on, though, is work. "He's one of the very few Silicon Valley people who really understands everything," Roux says. "Not just the technical, or work culture, but he's very financial. He was the only one of his peers who increased his holding in his company," which he did through buybacks.

If Ellison had flown a bit under the radar, it's probably because unlike most big Silicon Valley companies, Oracle has no consumer-facing businesses. Instead, it produces and sells an alphabet soup of B2B software, including ERP (enterprise resource planning), EPM (enterprise performance management), SCM (supply-chain management), and especially, at its core, DBMS (database management system).

Ellison co-founded Oracle as Software Development Laboratories in 1977 when Jimmy Carter was in the White House and Gen. Stansfield Turner was running the Central Intelligence Agency. The latter is salient because the name Oracle comes from the code name of a CIA-funded project that Ellison worked on. The CIA was Oracle's first customer, and it allowed Ellison's company to use the name.

In 1979, Oracle introduced the first commercial relational database, which essentially organizes and manages data sets, calling it Version 2 because he didn't think anyone would buy a Version 1. Oracle's databases ran on any hardware system, a decision that was really Ellison's first big Eureka moment and why Oracle came to dominate the database business.

Relational database software may sound prosaic, but in fact it's the central nervous system of any digital business, which now means every business -- a hugely powerful tailwind for Oracle. Competitors like Ingres, Ashton-Tate, Sybase, and Informix have come and gone, but Oracle remains the leader in this business, though SAP, IBM, and Microsoft also have significant shares.

Speaking of Microsoft, that company went public on March 13, 1986, a day after Oracle's initial public offering. Both stocks popped over 30%, and yet Microsoft grabbed more attention, as its application and operating software for PCs was easier for most of us to wrap our brains around. The rivalry with Gates and Microsoft has been more than just about stock prices and publicity, though, or even beyond personal -- Gates the nerd, Ellison the renegade. It has been technologically ideological. Gates believed that computing was about empowering the decentralized PC. Ellison argued that computing power should be centralized.

For decades, Ellison railed against Gates' take. ("I hate the PC with a passion," he said in 1996.) Turns out Ellison wasn't wrong. Though Microsoft is far bigger than Oracle -- with 4.9 times the revenue ($281 billion versus $57 billion) and 4.8 times the market cap ($3.85 trillion versus $823 billion) -- less than 20% of Microsoft's sales come from its PC segment (which includes Windows, Xbox, search, and news advertising).

In fact, both Microsoft and Oracle branched out from their roots. For Oracle, that meant first building applications on top of its database business to help customers connect and optimize their finance, sales, and supply-chain operations. Progress was slow, and in some cases Oracle's software was inferior to that of stand-alone competitors, but Ellison insisted on staying the course.

"He was very focused on organic growth and anti-acquisition," says Nelson. "His mantra was, 'We're not acquiring any technology.' And then three years later, he completely changed and went out and bought everything." The big spree started with a hostile takeover of PeopleSoft in 2005 for $10 billion, then Siebel Systems (run by frenemy Tom Siebel), BEA Systems, and more -- deals that put Oracle on the enterprise applications map.

Meanwhile, another seismic strategic shift was under way. Oracle's business had been based on selling what's known as on-premise or perpetual license software, where the company would install its software on customers' servers and computers, which then managed the products and paid a large upfront fee. But beginning in the late 1990s, it became evident that enterprise software could be cloud-based, hosted on the software vendor's servers, managed by the vendor, and paid for by subscription. This was called software as a service, or SaaS.

Oracle was slow to board this train, with Ellison calling cloud computing "complete gibberish" as late as 2009. By then, SaaS pioneer Salesforce, which first focused on web-based software for sales teams, had been around for a decade.

Behind the scenes, though, Ellison was hedging his bets. He personally invested in and served on the board of Salesforce. And he persuaded former Oracle executive Evan Goldberg to found NetSuite, a SaaS company that started off with web-based accounting software, which Ellison also invested in.

It took until 2010 or so for Ellison to move Oracle's applications into the cloud, and it wasn't until the middle of that decade that Oracle began to migrate its core database business, which came to be called Oracle Cloud Infrastructure, or OCI. By then, Amazon.com's AWS and Microsoft's Azure had built billion-dollar businesses, with market shares of 30% and more than 20%, respectively, while Alphabet's Google has 13%. OCI's share hovers in the low-single digits. Small wonder that Oracle's revenue growth slowed to a crawl in the 2010s -- from $35 billion in 2011 to $40 billion in 2020.

Now, though, the AI boom has given Oracle an enormous boost. For one thing, so much money has flooded into AI that it has jacked up all four cloud companies. "You only have a few players that can do this at scale, and Oracle's one of them, so by definition Oracle's benefiting," says Jefferies' Brent Thill. Oracle's stock has risen much more than the other three, in part because Oracle came from a lower revenue base.

Oracle also has some particular advantages. It has an integrated offering consisting of its revamped cloud service now powered by AI, its enterprise apps, and its latest core database product, called Version 23ai. Also, unlike Google and Microsoft, Oracle essentially has no competing AI customer-facing products: Instead of competing with large language models like OpenAI, it serves them, giving Oracle a Switzerland-like positioning.

And finally, Oracle has Larry, a consummate networker who last year regaled analysts with a tale about a dinner with Musk and Nvidia CEO Jensen Huang, where he begged the latter to "please take our money. No, take more."

Another impactful relationship has been with OpenAI CEO Sam Altman, highlighted initially in January with the unveiling of Stargate, a venture among Oracle, OpenAI, and tech investor SoftBank Group to spend up to $500 billion building data centers in the U.S. to help power OpenAI's development. The high-profile announcement was made with President Donald Trump on his second day in office. So far, the going has been slow on that front.

Flash-forward to September, a most eventful month for Oracle. First the company announced that longstanding CEO Safra Catz was transitioning to executive vice chair of the board and would be replaced by co-CEOs Clay Magouyrk and Mike Sicilia, who have a combined 28 years working at Oracle on cloud, AI, and applications.

More momentous was news that Oracle had increased its RPO -- that's "remaining performance obligations," not "run-pass option," football fans -- essentially the unfulfilled backlog of signed contracts, from $99 billion to $455 billion. (The revenue is to be recognized over five years.) This for a company with $57 billion in revenue for its fiscal year ending in May. No wonder Oracle's stock soared 36% the next day to a record closing high of $328. It has since settled back down to $297.

Oracle said three customers accounted for this future revenue, though some $300 billion of it came from a contract with OpenAI. That means Oracle's future prospects are tightly intertwined with a money-losing start-up with a current revenue run rate of just $10 billion. Moody's recently warned of Oracle's "counterparty risk," given that customer concentration. Competitors are raising their eyebrows, like the executive at a large rival software company who emailed me, "Let's wait and see what truly shapes out between now and 2029 on the RPO, and ultimately at what operating margin...;-)."

"The RPO is real and contracted, [but] contracts sometimes get amended," says Jackson Ader of KeyBanc Capital Markets. "Every marriage contract has the absolute intention to last until death, but it seems like there is at least some nontrivial likelihood that [the OpenAI RPO] will be amended."

As for the economics and margins of this new stream of revenue, Oracle will need mountains of cash, as in tens of billions of capital (much of it for Nvidia chips), to build out capacity to serve the business, says Jefferies' Thill. Oracle turned free-cash negative in fiscal 2025 for the first time since 1990, and analysts expect that to be the case for a number of years, as it plows dollars into the ramp-up.

With more than $90 billion of long-term debt, Oracle is more highly levered than its competitors. Though the company maintains its investment-grade debt rating, Moody's and S&P changed the company's outlook to Negative in July, when Oracle indicated that it had billions of dollars of new cloud revenue coming in the door and would have to increase spending. Late last month, the company issued $18 billion in bonds and will probably need to borrow billions more.

KeyBanc's Ader sees Oracle's cost of cloud revenue soaring from $10.9 billion in fiscal 2025 to $72.1 billion in 2029, pushing down its gross margin from 72% to 52%. Still, he has an Overweight rating and a $350 price target on the stock and sees net income surging from $17.3 billion this year to $42.6 billion in 2029. How does he square that? With a tremendous upswing in revenue, from $57.4 billion this year to an estimated $160.9 billion in four years.

As it turns out, Ellison's involvement with TikTok -- the giant Chinese social network with its addictive, AI-reliant recommendation algorithm, controlled by the Beijing-based technology company ByteDance -- is connected to this surging AI/cloud business. The story goes back to 2020, when Ellison, along with Walmart, made a bid to buy TikTok USA after the Trump administration attempted to force ByteDance to sell its U.S. operations. TikTok said that Oracle would run its secure cloud in the U.S., a coup for Oracle's then also-ran business. That deal never materialized, but in 2022 TikTok and Oracle formalized their relationship with "Project Texas," in which TikTok would migrate U.S. customer data onto Oracle's cloud.

And now in the latest act, Oracle, along with tech investor Silver Lake and MGX, an Abu Dhabi--owned investment firm, is looking to buy control of TikTok USA. Oracle would continue as TikTok's cloud provider and reportedly receive a licensed copy of its algorithm, which it would rework and monitor to ensure it's free from Chinese government influence or surveillance. The deadline for this deal to close, though it has previously been extended and must be greenlit by Trump, Congress, and the Chinese government, is Dec. 16.

"The TikTok deal doesn't stem from Larry saying, 'Hey, we should get into the content business,' " says someone who has worked closely with Ellison. "TikTok is important to Larry because it's a large customer. This is about creating a strategic bulwark around an important relationship. Imagine if TikTok gets sold to Microsoft? Also, Larry's doing something most CEOs couldn't do. After Time Warner swung and missed on AOL and gave up over half its business for an illusion, people might say, 'TikTok grew up really quickly. It could go away just as fast.' "

Ellison's other foray into the world of movies and TV through his son David's company Skydance Media is another side-door media play. David, who was mentored by Steve Jobs, first preferred acting, then producing, and founded Skydance in 2006. (A pilot, David named the company after what he calls a type of performing aerobatics.)

In August, Skydance bought Paramount -- which controls CBS as well the movie studio -- in a deal which valued the legacy media property at $8 billion, with $2 billion coming from RedBird Capital and $6 billion from the Ellison family. Safra Catz, who was previously on the board of Walt Disney, joined the board of the renamed Paramount Skydance.

Skydance is moving to make CBS less beholden to traditional media orthodoxy, as company executives might say. This past week, Skydance paid $150 million to buy the Free Press, the media company founded and run by journalist Bari Weiss, and made her editor in chief of CBS News, reporting to David Ellison.

Now Skydance, along with RedBird, is looking at far bigger game, specifically buying Warner Bros. Discovery, which could cost some $60 billion including assumption of long-term debt.

How much is Larry Ellison involved in Skydance, and how much is he funding? The answer to both is "significantly," but with limits. "Larry is emotionally involved in tennis and sailing, but not Hollywood," says a person familiar with the situation. "He sees these things more as data than movies." Still, Larry Ellison is on high-level deal calls and asks tough questions. How much of the $6 billion came for Paramount from Larry Ellison? "It's not as binary as Larry just funding David," says this person. "David is super smart and accomplished and has some of his own money. It's opaque, but clearly Larry puts in the bulk of it."

For a Warner Bros. Discovery deal, the Ellisons and RedBird may or may not turn to outside capital. Consider Larry Ellison's liquidity. He owns almost 1.2 billion shares of Oracle. The company pays a 50-cent quarterly dividend, so that's $2.3 billion a year. The company's Sept. 26 proxy notes that Ellison had pledged 346 million shares of his Oracle stock (worth about $100 billion) "as collateral to...secure personal term loans only used to fund outside personal business ventures."

Significantly, Skydance and RedBird intend to bring Larry Ellison's tech sensibility to create and distribute content though less-expensive, cloud-based alternatives to what legacy Hollywood and TV currently use. Note this point that David Ellison made recently: "I am confident that...CBS News and the Free Press will make big strides and be at the forefront of a much-needed transformation in how news is gathered, reported, and delivered." Stay tuned.

So, how long will Larry keep running the table -- or running at all? "Larry is fully engaged right now. He isn't going anywhere." That was what Steve Jobs told Fortune 25 years ago when Ellison was 56, though it seems he might have said it today, except that Jobs, of course, is gone. He died 14 years ago this past week. As for Gates, he left the field long ago, having stepped down from day-to-day duties at Microsoft in 2008. That leaves Ellison almost alone among the tech giants of his generation still in the game.

In that same Fortune article, Ellison himself said he was sticking around for "the duration," which has worked out pretty well so far. Stay tuned there, as well.

應版權方要求,你需要登入查看該內容

免責聲明:投資有風險,本文並非投資建議,以上內容不應被視為任何金融產品的購買或出售要約、建議或邀請,作者或其他用戶的任何相關討論、評論或帖子也不應被視為此類內容。本文僅供一般參考,不考慮您的個人投資目標、財務狀況或需求。TTM對信息的準確性和完整性不承擔任何責任或保證,投資者應自行研究並在投資前尋求專業建議。

熱議股票

  1. 1
     
     
     
     
  2. 2
     
     
     
     
  3. 3
     
     
     
     
  4. 4
     
     
     
     
  5. 5
     
     
     
     
  6. 6
     
     
     
     
  7. 7
     
     
     
     
  8. 8
     
     
     
     
  9. 9
     
     
     
     
  10. 10