# OpenAI, the empire burning $14 billion to rule the world

> OpenAI reportedly posted roughly $14 billion in losses for 2026, against approximately $13 billion in revenue, according to an analysis published in

- Author: Maxime Marquette
- Publisher: MadMax (https://mad-max.co)
- Section: Column
- Published: 2026-07-06
- Canonical URL: https://mad-max.co/en/article/openai-l-empire-qui-brule-14-milliards-pour-dominer-le-monde

## Key takeaways

- Reported 2026 losses of roughly $14 billion against about $13 billion in revenue, with internal projections of cumulative losses near $44 billion before profitability, possibly around 2029 or beyond.
- Revenue has tripled year over year, but infrastructure costs, including computing contracts and data centers, are rising faster.
- The IPO may be delayed to 2027, and OpenAI is reportedly considering offering a 5% stake to the US government.
- China's AI push and the US treatment of AI leadership as a national security issue frame the competitive stakes, while Anthropic is a direct rival.
- Most ChatGPT users pay nothing, and OpenAI has introduced advertising; Deutsche Bank analysts reportedly estimate cumulative losses could exceed $140 billion.

## Introduction: the dizzying price of the artificial intelligence race

### A number that makes your head spin

**OpenAI** reportedly posted roughly **$14 billion** in losses for 2026, against approximately **$13 billion** in revenue, according to an analysis published in early July 2026 that confirms internal documents already reported for months. This financial imbalance, alarming at first glance, illustrates the deliberate strategy of the company led by **Sam Altman**: spend massively today to dominate tomorrow's global market for **generative artificial intelligence**.

The company's internal projections anticipate **cumulative losses nearing $44 billion** before any realistic prospect of profitability, a horizon that, by some estimates, could stretch to 2029 or beyond. It's a figure that would have scared off any cautious investor, were it not for the near-religious conviction now animating the entire **Silicon Valley** ecosystem around the potential of **generative AI**.

### Revenue growth that still isn't enough

It would nonetheless be unfair to reduce **OpenAI** to a simple story of financial failure: the company has indeed tripled its revenue compared to the previous year, driven by explosive **ChatGPT** usage and a growing number of enterprise contracts. This spectacular growth, however, isn't enough to offset infrastructure costs rising even faster than revenue itself.

Chief Financial Officer **Sarah Friar** herself warned company leadership that revenue growth might not keep pace with the massive computing contracts already signed, a rare admission of vulnerability from a company accustomed to projecting an image of technological invincibility.

*I have to admit a certain fascination mixed with unease at these numbers. No company has ever burned this much cash while betting so openly on a future it doesn't fully control. It's either the boldest bet in tech history, or a bubble in the making.*

## The crushing weight of infrastructure costs

### Computing power, a bottomless financial pit

Most of **OpenAI's** losses stem from the colossal costs tied to the **computing infrastructure** needed to train and run its **artificial intelligence** models, expenses that include massive purchases of graphics chips, the construction of data centers, and the associated energy contracts. These costs, already considerable today, keep rising as models grow more complex and demand ever more computing power.

Some estimates put **inference costs** alone at nearly **$14 billion** for 2026 alone, an amount that rivals the company's total revenue and shows just how much the current business model of **generative artificial intelligence** rests on massive investments whose payoff remains largely uncertain.

### A structural dependence on technology partners

This cost explosion partly explains the complex and sometimes tense relationship between **OpenAI** and its main investor, **Microsoft**, which itself revealed that **OpenAI** had burned through **$12 billion** in a single quarter, a rate of capital consumption that worries even the partners most invested in the company's success.

This structural dependence on powerful financial and technology partners raises a fundamental question about **OpenAI's** real long-term autonomy, despite its status as a symbolic pioneer of the global **generative artificial intelligence** revolution.

*I honestly wonder: can we call it technological leadership when a company depends this heavily on outside capital just to keep operating? **OpenAI's** displayed dominance strikes me, at this stage, as far more fragile than its public image suggests.*

## The delayed IPO, an admission of caution or a calculated strategy

### A postponement to 2027 that puzzles observers

**OpenAI** is reportedly considering delaying its **initial public offering** to 2027, after watching the chaotic market debuts of several other highly anticipated tech companies. This caution, unusual for a company accustomed to projecting absolute confidence in its own future, could reflect real concern among leadership about how financial markets would react to losses this massive.

**OpenAI's** last known valuation stood at roughly **$852 billion**, a figure that seems hard to square with annual losses of **$14 billion**, unless one is betting that the future growth trajectory alone will justify this current stratospheric valuation.

### A surprising stake offered to the **US government**

In a move that is, to say the least, unexpected, **OpenAI** is reportedly considering offering a **5% stake** to the **US government**, a maneuver that could both secure valuable political backing in **Washington** and get ahead of potentially stricter regulation of the **artificial intelligence** sector in the United States.

This proposal comes as the **Trump administration** closely examines national security questions tied to **artificial intelligence**, a context in which such a government stake could serve as a strategic shield for **OpenAI** against possible future restrictions.

*I find this idea of a government stake revealing of a changing era: artificial intelligence is no longer just a matter for visionary entrepreneurs, it's becoming a national security issue that states now want to control directly.*

## Chinese competition, a systemic threat to the Western lead

### A technological race with no zero-sum outcome

While **OpenAI** burns billions to maintain its technological edge, **China** is investing massively in its own **artificial intelligence** capabilities, with the openly declared ambition of catching up to, then overtaking, American companies in this strategic field. This technological race goes far beyond a simple commercial stake: it directly shapes the global geopolitical balance for decades to come.

The **US government** now explicitly treats maintaining America's technological lead in **artificial intelligence** as a top-tier **national security** matter, a perspective that partly explains authorities' tolerance for the massive losses of companies like **OpenAI**, deemed strategically indispensable.

### The **West** cannot afford to lose this race

This technological competition between the United States and **China** puts the entire Western **artificial intelligence** ecosystem under constant pressure to innovate rapidly, even if that means accepting levels of financial loss that would be judged unacceptable in any other traditional economic sector.

The risk for the **West** lies not only in a possible financial failure at **OpenAI**, but in the possibility that this frantic race for capital could weaken the entire American technology ecosystem against Chinese rivals who benefit from more direct and stable state support over the long run.

*I deeply believe the **West** cannot afford to lose this technological race against **China**, but I worry that the financial fragility of players like **OpenAI** could paradoxically become a strategic vulnerability rather than an asset.*

## **Anthropic** and the battle of direct rivals

### Internal competition within the American ecosystem

**Anthropic**, **OpenAI's** chief direct rival in the American **generative artificial intelligence** market, recently launched **Claude Science** with the explicit goal of broadening its own revenue streams ahead of its own IPO. This rivalry between American AI giants illustrates an intense market dynamic that could, over time, accelerate technological innovation while simultaneously worsening financial pressure across the entire sector.

This competition between **OpenAI** and **Anthropic** reflects an American **artificial intelligence** ecosystem that is both dynamic and fragile, where several major players are simultaneously burning through considerable sums in hopes of establishing lasting dominance in a market still very much under construction.

### A market that might not have room for everyone

Some analysts are openly questioning whether the **generative artificial intelligence** market can financially sustain all the players currently competing, a concern that could, over time, lead to a brutal sector-wide consolidation should investors lose patience with the prolonged absence of profitability.

This prospect of consolidation, though speculative at this stage, deserves to be taken seriously by all sector observers, given how unsustainable current spending levels appear over the very long term absent a major technological breakthrough justifying these investments.

*I think the generative AI market is currently going through a phase where collective euphoria is masking real financial fragility. I'm not predicting a collapse, but I refuse to give in to the blind enthusiasm I too often see in media coverage of this sector.*

## Users, the silent hostages of the business model

### A paying-user ratio that remains too low

Despite the hundreds of millions of weekly users **ChatGPT** claims, an overwhelming proportion of those users bring no direct financial contribution to the company, an imbalance that largely explains why revenue growth is struggling to keep pace with the platform's otherwise impressive growth in active users.

This reality poses a central question for the future of **OpenAI's** business model: how to convert a massive but mostly free user base into a revenue source sufficient to justify the colossal investments already committed to the **computing infrastructure** needed to run its services.

### The lure of advertising, a risky paradigm shift

**OpenAI** recently introduced **advertising** within **ChatGPT**, a decision marking a significant strategic pivot for a company that had initially positioned itself as pursuing a selfless mission in service of humanity, before having to reckon with the unforgiving economic realities of its own growth model.

This shift toward advertising, while understandable in strictly financial terms, raises legitimate questions about the long-term trajectory of a company that will likely need to keep adjusting its business model under the constant pressure of **technology infrastructure** costs.

*I stay clear-eyed on this point: **OpenAI's** original idealism, that of artificial intelligence serving the common good, is gradually giving way to the financial imperatives of a company that has, in effect, become a commercial machine like any other.*

## Donald Trump's ambiguous role in the rise of AI

### An administration that supports American innovation

The **Trump administration**, despite its repeated criticism of other parts of the American tech ecosystem, appears broadly supportive of maintaining American supremacy in **artificial intelligence**, a sector viewed as strategic for the future economic and military competitiveness of the **United States** against **China**.

This stance, consistent with **Trump's** broader push to strengthen American technological power against international rivals, could smooth negotiations over the government stake being considered in **OpenAI's** capital.

### Internal tensions over the release of the most advanced models

Reports have revealed that the **Trump administration** asked **OpenAI** to stagger the release of information about its newest models, a direct government intervention that illustrates the growing geopolitical sensitivity surrounding the most cutting-edge advances in **artificial intelligence**.

This direct government involvement in the commercial decisions of a private company, however strategic it may be, deserves close scrutiny, as it could set a significant precedent for the entire American technology sector in the years ahead.

*I credit the Trump administration with grasping the major strategic stakes artificial intelligence represents against **China**, but I remain wary of any excessive government interference in private company affairs, even for legitimate national security reasons.*

## Growing skepticism among financial analysts

### Loss projections that keep getting worse

Several financial analysts, including those at **Deutsche Bank**, estimate that **OpenAI's** cumulative losses could exceed **$140 billion** before achieving any lasting profitability, a figure far above the **$44 billion** initially mentioned in the company's internal documents, reflecting a gradual deterioration of financial prospects as the months go by.

This constant upward revision of loss projections is fueling growing skepticism among some institutional investors, who legitimately question the long-term sustainability of a business model built on promises of profitability that keep getting pushed further into the future.

### The ultimate bet on an uncertain future profitability

**OpenAI's** leadership nonetheless continues to voice confidence in future profitability, anticipated around 2029 or 2030 depending on the scenario, a horizon distant enough to reassure some investors while leaving real uncertainty over the company's actual ability to keep that promise.

This bet on the future, bold as it is, rests largely on the assumption that global demand for **generative artificial intelligence** will keep growing fast enough to retroactively justify investments that a significant portion of financial observers in the sector currently consider excessive.

*I'm not trained as an economist, and I own that limitation: but plain common sense urges caution in the face of profitability projections that keep getting pushed back. Financial history is full of companies that promised a bright future that never arrived.*

## The **data centers**, an invisible but colossal infrastructure

### An unprecedented construction race

The construction of **data centers** meant to power **OpenAI's** models represents an infrastructure investment unmatched in the history of technology, with financial commitments cited at roughly **$600 billion** through 2030, according to some specialized analysts' estimates.

This frantic race to build massive physical infrastructure raises legitimate questions about the associated energy consumption, an environmental issue increasingly scrutinized by observers even as the race for **artificial intelligence** keeps intensifying worldwide.

### An energy challenge that goes beyond the financial question alone

This energy dimension of **artificial intelligence** development represents a broader societal challenge that goes far beyond the purely financial considerations usually discussed around **OpenAI**, raising structural questions about whether Western power grids can absorb such rapidly growing demand.

This energy reality, often overlooked in public debate focused solely on the performance of **artificial intelligence** models, deserves far more sustained political and industrial attention from Western governments in the years ahead.

*I think we're collectively underestimating the scale of the energy challenge posed by this race for artificial intelligence. There's plenty of talk about algorithms and **billions of dollars**, and far too little about the power plants that will need to be built to run all of it.*

## Ethical concerns too often pushed to the back burner

### Financial pressure that could erode technological caution

This constant financial pressure raises fears among some observers that **OpenAI's** commercial imperatives are gradually taking precedence over the ethical and safety considerations that had nonetheless guided the company's founding, originally conceived as a nonprofit project before its later transformation.

This structural shift, from idealistic research lab to commercial enterprise under constant financial pressure, illustrates a fundamental tension running through the entire **generative artificial intelligence** sector worldwide.

### The vigilance needed against a headlong race

This ethical vigilance remains all the more necessary since international competition, particularly against **China**, could push Western players to accelerate the deployment of technologies that are still insufficiently tested, at the risk of sacrificing caution on the altar of competitive speed.

This tension between speed of innovation and ethical caution is, in my view, one of the most underestimated issues in the current race for **generative artificial intelligence**, an issue that deserves far more sustained attention from Western regulators.

*I'm genuinely worried that the race for billions is overtaking the caution needed for such a transformative technology. History will not forgive our generation for having sacrificed safety on the altar of competitive speed.*

## The talent race and the salary war in **Silicon Valley**

### Staggering pay to retain key researchers

**OpenAI** is waging a fierce battle to retain its most talented **artificial intelligence** researchers, offering compensation packages that sometimes reach tens of millions of dollars over several years, a salary inflation directly fueled by fierce competition with **Anthropic**, **Google DeepMind**, and Chinese tech giants aggressively recruiting from the same narrow global talent pool.

This salary war weighs heavily on the company's cost structure, adding further pressure to finances already strained by massive investments in **computing infrastructure**, at a time when every departure to a direct competitor potentially represents a major loss of strategic advantage for **OpenAI**.

### An executive exodus that worries sector observers

Several departures of senior executives and senior researchers have been reported in recent months, some joining direct competitors, others launching their own specialized **startups**, a phenomenon fueling questions about the company's internal stability at a particularly critical moment in its financial development.

These successive departures raise a fundamental strategic question for **OpenAI's** future: how to maintain a coherent culture of innovation and a lasting technological edge when top talent is constantly courted by ever more generous financial offers from Western and Chinese competitors alike.

*This point strikes me as underestimated in current media coverage: there's plenty of talk about the billions lost on infrastructure, but very little about this silent talent drain that, over time, could cost **OpenAI** far more than any server bill.*

## The impact on the artificial intelligence startup ecosystem

### A dreaded domino effect on young companies dependent on OpenAI's APIs

Thousands of **startups** across the Western world are building their products directly on **OpenAI's** **APIs**, a strategic dependence that worries an increasing number of investors as the parent company's financial troubles become public and questions about its long-term viability multiply in the trade press.

Any future decision on **pricing restructuring** or a shift in commercial policy could have cascading repercussions across this entire ecosystem of young innovative companies, revealing just how much the concentration of technological power around a handful of dominant players represents a systemic risk largely underestimated by the market.

### **Venture capital** investors rethink their funding strategies

Several **Silicon Valley** venture capital funds are now revising their investment criteria for **artificial intelligence** startups, demanding greater diversification of language model providers rather than exclusive dependence on **OpenAI**, a newfound caution reflecting awareness of the risk of excessive concentration.

This diversification directly benefits competitors like **Anthropic**, **Mistral AI** in Europe, or open-source solutions, illustrating how one dominant player's financial troubles can paradoxically strengthen the overall resilience of the Western **artificial intelligence** ecosystem over the medium term.

*This may be the most positive consequence of this crisis: a forced diversification that, if well managed, could strengthen rather than weaken the West's technological resilience against Chinese competition, by avoiding an all-in bet on a single player that has become financially fragile.*

## Financial markets' reaction to persistent uncertainty

### A valuation that defies traditional financial logic

Despite announced losses of **$14 billion**, **OpenAI's** valuation keeps climbing on private secondary markets, a paradoxical situation that reminds some analysts of the excesses seen during the **dot-com bubble** of the early 2000s, when the promise of future growth justified valuations disconnected from present financial results.

This disconnect between actual financial performance and anticipated market valuation illustrates the fundamental difficulty of correctly valuing a technology as disruptive as **generative artificial intelligence**, whose long-term economic transformation potential remains largely speculative despite the already colossal investments made across the entire sector.

### Dissenting voices emerge among major institutional investors

Some large institutional investment funds are beginning to publicly voice reservations about the sustainability of **OpenAI's** current growth model, demanding greater transparency about the precise path to profitability before committing additional capital in the company's next funding rounds.

This growing institutional caution could, if it spreads, seriously complicate **OpenAI's** ability to raise the tens of billions of additional dollars needed to fund its continued expansion strategy, at a time when international competition leaves no room for strategic slowdown.

*I watch this growing market skepticism with a mix of caution and relief: it was about time financial logic caught up with technological euphoria, though I sincerely hope this correction won't hold back the legitimate innovation the West needs against its strategic rivals.*

## What this crisis reveals about the entire tech sector

### A window into the excesses of the current speculative bubble

**OpenAI's** financial troubles, even as the company remains undeniably a technological leader in its field, act as a broader window into the potential excesses of a speculative bubble that has built up around **generative artificial intelligence** in recent years, driven by optimism sometimes disconnected from concrete financial realities.

This question goes far beyond **OpenAI** alone: it calls into question the overall sustainability of an entire technology sector that has raised colossal sums based on growth promises that remain, for now, largely theoretical and not yet financially realized.

### A lesson in caution for the entire Western ecosystem

This situation imposes a collective lesson in caution on the entire Western technology ecosystem: innovation, however promising, cannot indefinitely escape fundamental economic laws without eventually running up against an unavoidable financial reality.

This caution should not, however, turn into excessive timidity: the **West** cannot afford to slow its race in **artificial intelligence** against Chinese competition, which faces neither the same market constraints nor the same short-term shareholder pressure.

*I'll close this reflection with a deliberate paradox: we must both invest massively to avoid losing the technological race against China, and stay clear-eyed about the real financial risks this race is placing on the entire Western innovation ecosystem.*

## Conclusion: a colossal bet whose outcome remains uncertain

### A company at a crossroads

**OpenAI** now stands at a decisive strategic crossroads: keep burning billions of dollars in hopes of future technological dominance, or gradually adjust its business model to regain a more sustainable financial trajectory over the medium term. This choice will determine not only the company's own future, but also, to a certain extent, the overall trajectory of the Western **generative artificial intelligence** ecosystem.

The coming months look decisive for clarifying this trajectory, amid a likely delay of the **IPO**, a possible government stake, and growing competitive pressure from both **Anthropic** and Chinese **artificial intelligence** players.

### A stake that extends far beyond OpenAI alone

Beyond **OpenAI's** own fate, this situation illustrates the deep tensions running through the entire technological revolution underway: how to sustainably finance innovation this costly, without sacrificing the caution needed for a technology whose societal implications remain largely underestimated by the general public.

The outcome of this colossal financial bet will largely determine whether the **West** manages to keep its technological lead over **China**, or whether this headlong rush for capital ends up weakening the very players it was meant to strengthen on the international stage.

*I'll end this column with deliberate caution: I don't know whether OpenAI will win its bet. But I do know that the outcome of this financial and technological battle will have consequences reaching well beyond **Silicon Valley** alone, into the global geopolitical balance of the decades ahead.*

**By Maxime Marquette, columnist**

## Columnist's transparency note

### Editorial stance

I am not a journalist but a **columnist and analyst**. My expertise lies in observing and analysing the geopolitical, economic and strategic dynamics that shape our world. My work is to take political strategies apart, to understand global economic movements, to put the decisions of international actors in context and to offer analytical perspectives on the changes that are redefining our societies.

I do not claim the cold objectivity of traditional journalism, which stops at reporting facts. **I claim analytical clarity**, rigorous interpretation and a deep understanding of the complex issues that concern us all. My role is to give meaning to the facts, to place them in their historical and strategic context, and to offer a critical reading of events.

### Method and sources

This text keeps the basic distinction between **verified facts and interpretive analysis**. The rule never changes: a factual claim is published only when a verifiable source supports it, and the sources this article actually used are listed under **Sources**, never here.

**Primary source families** used by the publication, when the story has them: official statements from governments and international institutions, public statements by political leaders, reports from intergovernmental organisations, dispatches from recognised international news agencies.

**Secondary source families**: specialised publications, internationally recognised news media, analyses from established research institutions, reports from industry organisations.

When an article cites statistical, economic or geopolitical data, it comes from the institutions that produce it (intergovernmental organisations, central banks, national statistics offices), and the exact institution is named under **Sources**.

### Nature of the analysis

The analyses, interpretations and perspectives in the analytical sections of this article are a **critical, contextual synthesis** based on the information available, the trends observed and the expert comments cited in the sources consulted.

My role is to interpret these facts, to set them in the context of today's geopolitical and economic dynamics, and to give them a coherent meaning within the larger story of the changes shaping our time. These analyses reflect expertise built through continuous observation of international affairs and an understanding of the strategic mechanisms that drive global actors.

This article describes a documented state of affairs on its publication date, not a prediction: later developments may shift its outlook. No update is promised in advance; when an article is corrected or completed, the change is dated in the text.

## Sources

### Primary sources

- Weekend Dispatch — analysis of OpenAI's finances, July 4, 2026

- OpenAI's 2026 Crisis: Missed Targets, $14 Billion Losses — Powered Magazine

### Secondary sources

- Anthropic launches Claude Science ahead of IPO — Yahoo Finance

- OpenAI may delay its IPO to 2027 — Forbes, June 25, 2026

- OpenAI considers giving Trump administration 5% stake — The Independent, July 2, 2026

- OpenAI's revenue, growth estimates fall short as company races toward IPO — CNBC, April 28, 2026

---

Source: Maxime Marquette, MadMax. When citing, name the author and link https://mad-max.co/en/article/openai-l-empire-qui-brule-14-milliards-pour-dominer-le-monde