Strip away the superlatives and the core fact is still staggering: OpenAI has closed a $110 billion funding round, the largest private capital raise in technology history, with cheques from Amazon, NVIDIA and SoftBank. The deal values the ChatGPT maker at $730 billion on a pre-money basis, or $840 billion including the fresh capital. For context, that puts it ahead of many ASX-listed companies combined.
The breakdown, as reported by Engadget and confirmed by CNBC, is $50 billion from Amazon, $30 billion from NVIDIA, and $30 billion from SoftBank. The round remains open, with OpenAI expecting more investors to join. Sovereign wealth funds and other institutional players have reportedly been in discussions about contributing an additional $10 billion.
A Deal Built on Circular Logic
Look past the headline number and the structure of these partnerships reveals something worth scrutinising. Amazon is putting in $50 billion, but OpenAI has simultaneously committed to consuming at least 2 gigawatts of Amazon's Trainium compute capacity, its custom-designed AI training accelerator. AWS has also been named the exclusive third-party cloud distribution provider for OpenAI Frontier, the company's agentic enterprise platform. In real terms, this translates to: Amazon writes a cheque to OpenAI, and OpenAI routes a significant portion of it back through Amazon's cloud infrastructure.
The NVIDIA arrangement follows the same pattern. NVIDIA is investing $30 billion while OpenAI has pledged to consume 2 gigawatts of training capacity on NVIDIA's next-generation Vera Rubin systems, plus an additional 3 gigawatts of compute resources for AI inference tasks. The chip giant gains a financial stake in one of its largest customers, tightening a relationship that was already deeply intertwined.
This circular capital flow is not necessarily irrational. Infrastructure commitments of this scale require both parties to share long-term risk. But investors watching from the outside should understand that the gross investment figures do not translate one-for-one into freely deployable cash.
Amazon's Conditions: Including, Potentially, the End of the World As We Know It
Of Amazon's $50 billion commitment, only $15 billion is unconditional. The remaining $35 billion will flow only once certain milestones are met. According to Reuters, one reported condition is that OpenAI achieves artificial general intelligence, the theoretical threshold at which AI systems match or exceed human-level cognitive ability across all domains.
That condition deserves a moment's pause. AGI remains one of the most contested concepts in computer science. OpenAI CEO Sam Altman once predicted it would arrive in 2025 but has since stepped back from the term altogether. Many researchers believe the concept is either decades away or fundamentally ill-defined. Tying $35 billion to a milestone that may never arrive, or whose arrival no one can agree on, is an unusual structuring choice, to put it diplomatically.
The Losses Are Real. So Is the Growth.
Here's what the financial trajectory actually tells us: OpenAI lost approximately $5 billion in 2024. Estimates place the 2025 loss at around $8 billion. The company's own internal forecasts project a $14 billion loss in 2026, with cumulative losses from 2023 through 2028 expected to reach $44 billion. The company does not project profitability until 2029, when it forecasts $100 billion in annual revenue.
Those who are sceptical of that 2029 target have legitimate grounds for concern. OpenAI's revenue was approximately $3.7 billion in 2024 on a reported loss of $5 billion. Reaching $100 billion in revenue within four years would require growth that has few precedents in enterprise software history. The most bullish comparable is NVIDIA itself, which generated around $130 billion in revenue in 2025 off the back of what many describe as the largest hardware boom in modern tech history. OpenAI is betting it can replicate that scale through software and services.
Yet the user growth story provides some genuine substance to those projections. ChatGPT now serves more than 900 million weekly active users globally, with over 50 million paying consumer subscribers and more than 9 million enterprise business users. The platform has embedded itself into workflows at a speed few software products have matched. For investors, the signal is clear that the demand side is not the problem. The question is whether the cost side can ever be tamed.
What the Market Has Not Yet Priced In
Tech stocks have had a difficult start to 2026, with investors growing warier about whether AI expenditure will generate returns commensurate with the capital committed. NVIDIA's own share price was punished this week after the company confirmed it would direct money into the broader AI ecosystem rather than returning it to shareholders through buybacks. That is a meaningful signal about where market sentiment sits relative to corporate confidence in the AI investment case.
The bulls argue this is exactly the kind of moment where long-term positioning matters. Amazon CEO Andy Jassy said in a statement that OpenAI would be "one of the very big winners" over the long term, and that AWS was well placed to support that growth. That view is consistent with the logic that whoever controls AI infrastructure at scale will capture disproportionate value as the technology matures.
The sceptics point to the widening losses, the circular capital arrangements, and the fact that OpenAI's existing partnership with Microsoft, its longest-standing major backer, was not extended or expanded in this round. Microsoft did not participate, though OpenAI insisted the terms of that relationship remain unchanged. The absence is notable even if it is not conclusive.
Critically, OpenAI's governance structure adds another layer of complexity. The company operates under a capped-profit model with a non-profit parent. Fresh capital of this size will intensify pressure to convert that structure into something more conventional, particularly if an IPO, which is widely anticipated this year, moves forward. The Reserve Bank of Australia and local regulators will be watching the downstream effects on Australian technology investment flows, even if the action is happening squarely in Silicon Valley.
What is genuinely hard to dispute is that the scale of commitment here, from three of the world's most capital-disciplined technology companies, reflects a considered bet that AI is not merely a product cycle but a generational infrastructure shift. Whether OpenAI is the primary beneficiary of that shift, or simply the most expensive toll booth on the road to it, is a question investors have committed $110 billion to answering.