On Friday, Netflix announced it would acquire Warner Bros. Discovery’s studio and streaming assets for $72 billion. The deal brings a century-old library under Netflix’s roof: Harry Potter, Batman, Game of Thrones, The Sopranos, Friends. The entertainment press responded with predictable narratives. Streaming consolidation. The death of Hollywood. The triumph of the disruptor.
These narratives are not wrong. They are incomplete.
The more interesting question is not what Netflix bought, but what it is defending against. And the answer has less to do with Hollywood than with a company 350 miles north of it.
The Surface Logic
The deal makes conventional sense. The streaming market has fractured into too many subscriptions chasing too few households. HBO Max, Paramount+, and Peacock have struggled to reach scale. Consolidation was inevitable. Netflix, cash-rich and dominant, struck when Warner Bros. Discovery was weakest.
The financial case is straightforward: overlapping subscribers, redundant infrastructure, $2-3 billion in annual cost savings by year three. Netflix expects the deal to be accretive to earnings per share by year two.
But Wall Street remains divided. Barclays analysts wrote that it remains “unclear what problem or opportunity Netflix is solving for that couldn’t have been achieved organically.” Dave Novosel, senior bond analyst at Gimme Credit, called the Ebitda multiple of more than 25x “extravagant,” noting that Netflix will assume nearly $11 billion of debt and require a $59 billion bridge loan (one of the largest in corporate history) to finance the transaction.
That skepticism is reasonable if the deal is primarily about content. It becomes less reasonable if Netflix is playing a different game entirely.
The Deeper Logic
Melissa Otto, head of research at S&P Global Visible Alpha, offered a different interpretation in an interview with Fortune. “I think there’s this much bigger conversation that is being missed,” she said. “Google and its TPU chips.”
She is not alone in this assessment. The infrastructure dynamics support her thesis.
Google’s tensor processing units have become exceptionally capable at generating video. The company released its seventh-generation TPU, Ironwood, in November 2025, offering 10x peak performance improvement over TPU v5p and more than 4x better performance per chip for both training and inference workloads. Anthropic has committed to accessing up to 1 million of these chips to run its Claude models. Google must now double its AI serving capacity every six months to meet demand.
YouTube already commands a dominant position in U.S. television viewing. According to Nielsen’s July 2025 Media Distributor Gauge, YouTube captured 13.4% of TV watch-time, establishing the largest lead by any media company since Nielsen began tracking in November 2023. Netflix held 8.8%. The gap is not closing. YouTube’s share of TV viewing has grown from 7.9% to over 13% in two years, with viewing from adults aged 65+ nearly doubling in that period.
The convergence is visible. Google’s Veo 3, released in May 2025, generates near-cinematic quality video with synchronized audio (dialogue, sound effects, and ambient noise) in a single pass. DeepMind CEO Demis Hassabis described the release as “the moment when AI video generation left the era of the silent film.” By October 2025, tens of millions of videos had been generated through the platform.
If this trajectory continues, the cost of producing high-quality video could collapse. Netflix’s historic advantage (its ability to outspend competitors on original content) would erode. Anyone with access to the right chips could generate entertainment.
Netflix appears to have recognised this threat. Its response is not to compete on production infrastructure, which it cannot win, but to acquire something Google cannot easily replicate: intellectual property at scale.
Intellectual Property as Training Data
The traditional value of IP lies in franchise rights, merchandising, and sequels. Warner Bros. has been extracting value from Batman for over eight decades. But in an AI-driven future, IP may serve a different function.
Otto raised an arresting scenario: imagine generating new episodes of Columbo starring an AI-rendered Peter Falk, who died in 2011. The style, rhythm, and character of the original would be preserved, but the content would be new. To do this legally and effectively, one would need access to the original material. Not just the rights, but the raw footage, the visual language, the corpus.
This reframes what Netflix is acquiring. Warner Bros. is not merely a library of films and television series. It is a dataset. Every frame of Harry Potter, every scene transition in The Sopranos, every lighting choice in Casablanca represents training material for future AI systems. The value is not only in what audiences will watch, but in what machines will learn.
Netflix increases “the volume and diversity of content it can feed into recommendation systems, experimentation, and, eventually, its own AI-driven video tools” through this acquisition, as Otto noted. The deal also brings potential exposure to Warner Bros.’ $6-7 billion in annual advertising revenue, an area where Google’s Alphabet has dominated and where Netflix has only been active since 2022.
Netflix is not buying Batman. It is buying every pixel Batman has ever occupied on screen.
Alternative Motivations and Structural Considerations
The AI thesis does not preclude other strategic rationales. Several factors likely contributed to Netflix’s calculus.
International rights consolidation. Warner Bros. holds distribution rights across 170+ territories for decades of content. In an era of fragmented licensing, owning the underlying rights simplifies global rollout and reduces ongoing negotiation costs.
Theatrical distribution infrastructure. Warner Bros. maintains theatrical release agreements through 2029 and relationships with major cinema chains worldwide. Netflix has historically struggled with theatrical distribution. The acquisition provides instant infrastructure without the reputational cost of building it independently.
Debt structure as strategic instrument. The $59 billion bridge loan and assumed $11 billion debt create a financing structure that few competitors could replicate. This is not merely a cost but a barrier. Paramount Skydance, despite backing from the Ellison family, could not match the terms.
Creator relationships. HBO’s reputation for creative freedom has attracted prestige talent for decades. The Sopranos, The Wire, Game of Thrones, and Succession were built on relationships that took years to cultivate. Netflix inherits these relationships and the institutional knowledge behind them.
These factors do not contradict the AI thesis. They complement it. Netflix is acquiring a multi-dimensional asset whose value compounds across distribution, rights, relationships, and training data simultaneously.
The Corpus War
The competitive landscape is now defined by three distinct models.
Google controls production infrastructure (TPUs), distribution (YouTube), and monetisation (advertising). YouTube’s ad revenue reached $8.93 billion in Q1 2025, up more than 10% year-over-year. Its implicit bet is that video creation will become cheap and abundant, and that economic value will accrue to whoever controls discovery and advertising around that content. Sundar Pichai has acknowledged that compute capacity remains the bottleneck: when Veo launched, Google “would have gotten more users but we just couldn’t because we are at a compute constraint.”
Netflix controls premium content libraries, subscriber relationships, and recommendation systems. Its bet is that audiences will continue to pay for quality, and that premium IP cannot be commoditised. By acquiring Warner Bros., it is deepening that moat. Co-CEO Ted Sarandos framed it plainly: “Our mission has always been to entertain the world. Together, we can give audiences more of what they love and help define the next century of storytelling.”
TikTok controls attention allocation and creator economics. Its bet is that user-generated content, intelligently distributed, can capture more engagement than professionally produced entertainment. Otto noted that TikTok remains “a formidable competitor” in this landscape.
Each model represents a different theory of where value will concentrate as AI transforms media production. Google believes in infrastructure. Netflix believes in corpus. TikTok believes in algorithmic distribution.
As Otto put it: “In a world where TPUs can make high-quality video basically for free, any player lacking both the chips and the content could find itself outgunned.”
Netflix’s $72 billion is a declaration: in the coming war over video, owning the raw material matters more than owning the factory.
Regulatory Risk and Political Dimensions
The deal is not yet closed. It requires regulatory approval and is expected to complete in the third quarter of 2026. The U.S. Department of Justice will review it, and Netflix’s combined streaming share may exceed the 30% threshold that typically triggers antitrust scrutiny.
Netflix has prepared its defence. It will argue that 75% of HBO Max subscribers already subscribe to Netflix, making the services complementary rather than competitive. It will further argue that YouTube and TikTok should be included in any market definition, which would dilute Netflix’s apparent dominance.
The political dimension complicates matters. Paramount Skydance, backed by Larry Ellison (the world’s second-richest man) and his son David Ellison, lost the bidding war. The Ellison family has “longstanding close ties to President Donald Trump,” whose administration praised their earlier acquisition of Paramount. Their description of the sale process as “tainted” signals potential lobbying efforts against approval.
Representative Darrell Issa (R-Calif.) raised concerns in a November letter to Trump administration officials, writing that Netflix “currently wields unequaled market power.” Senator Elizabeth Warren has also objected. The Hollywood Writers Guild urged that the deal “must be blocked.”
Netflix has included a $5.8 billion breakup fee if regulatory bodies block the transaction. The company is clearly pricing in uncertainty.
But even if the deal fails, the signal has been sent. Netflix has identified Google, not Disney or Amazon, as its primary long-term threat. It has reframed intellectual property as strategic infrastructure. It has placed its bet on corpus over compute.
What Comes Next
The AI video production wave has not yet arrived in mainstream form. Generative tools remain expensive, inconsistent, and legally uncertain. The recent Hollywood strikes made AI-generated likenesses a central issue, and the legal frameworks remain unsettled. But the trajectory is clear. Production costs will fall. Output will increase. The bottleneck will shift from creation to curation, and to ownership of the underlying material.
Otto suggested that the future is unpredictable but the opportunity is real: “It’s kind of exciting because it means that it’s anybody’s game.” She acknowledged that entertainment lawyers who wrangle over likeness rights “may be the real story” in the coming years.
Netflix is positioning itself for that shift. By acquiring Warner Bros., it secures not just a content library but a defensible position in the coming corpus war. Whether the price was right remains to be seen. Whether the thesis is correct will take longer to determine.
What is already visible is the nature of the bet. Netflix did not pay $72 billion to win the streaming wars. It paid $72 billion to build a moat against Google’s TPU + AI + YouTube triad.
The arms race over video corpus continues.
Sources
- Fortune, “Netflix to acquire Warner Bros.’ studios and HBO Max in landmark $72 billion deal,” December 5, 2025
- Fortune, “Top analyst says Netflix’s $72 billion bet on Warner Bros. isn’t about the ‘death of Hollywood’ at all. It’s really about Google,” December 5, 2025
- Google Cloud Blog, “Ironwood TPUs and new Axion-based VMs for your AI workloads,” November 2025
- CNBC, “Google must double AI serving capacity every 6 months to meet demand,” November 21, 2025
- Nielsen, “YouTube, Netflix Ride the Wave of Summer Streaming Highs in Nielsen’s Media Distributor Gauge,” August 26, 2025
- Nielsen, “YouTube Achieves Best Monthly Performance to Date,” March 25, 2025
- Google Cloud Blog, “Veo 3 available for everyone in public preview on Vertex AI,” June 26, 2025
- Wikipedia, “Veo (text-to-video model)”
- Apatero Blog, “Google Veo 3.1 Complete Guide,” December 2025
- Variety, “Netflix to Buy Warner Bros. and HBO Max in $82.7 Billion Deal,” December 5, 2025
- The Keyword, “YouTube dominates the streaming market, surpassing Netflix,” July 2, 2025
- Fortune, “Netflix–Warner Bros. deal sets up $72 billion antitrust test,” December 5, 2025
- Bloomberg via Fortune, “Hollywood writers say Warner takeover ‘must be blocked,'” December 5, 2025
- Associated Press, “Hollywood AI strike,” 2023
Disclaimer: This article is a market analysis and strategic commentary, not investment advice. It synthesizes publicly available information from various news sources, analyst reports, and company statements. The interpretations and conclusions presented reflect the author’s analysis of reported facts and expert opinions. Readers should conduct their own research and consult qualified professionals before making any investment or business decisions.