Nvidia CFO Colette Kress has made an ambitious prediction: its AI labs — companies that build the most advanced artificial intelligence models — will become the largest technology companies in history. The conclusion is based on the company’s financial data and plans for both its products and infrastructure investments.
Kress emphasizes that these „frontier AI labs“ are already established companies with rapidly growing customer adoption. This is borne out by Nvidia’s order book figures: the company has $500 billion in orders for its Blackwell and Rubin architectures through 2026, and those numbers don’t include any OpenAI activity. OpenAI’s collaboration with Nvidia is complementary, and its scale could significantly expand that number.
Huge investments in AI infrastructure
CEO Jensen Huang estimated that OpenAI’s current and future deployment projections could require about 12 GW of computing power, potentially growing to 16 GW. In dollar terms, this power demand equates to about $600 billion in Nvidia computing resources by 2030. Kress also points out that Nvidia is not just selling processors, it is also financing the physical infrastructure needed for those processors. For example, Nvidia is investing $1.5 billion in SB Energy to support OpenAI’s 20-year lease on the PORTS-Pike data center in Ohio, which will have an initial capacity of 4.25 GW, one of the largest in the world.
Nvidia is also providing credit support and aligning a broader investment framework with OpenAI that could value it at up to $100 billion. While the framework hasn’t been finalized until the end of 2025, its size demonstrates how closely the companies are connected.
Global AI infrastructure spending up to $4 trillion
Kress predicts that global AI infrastructure investment will reach $3 trillion to $4 trillion by the end of the decade. That includes not only the big cloud platforms like Microsoft, Google, and Amazon, but also the AI labs themselves. Nvidia is already reaping a significant portion of that spending: In the second quarter of 2027, the company reported $96.2 billion in revenue, with $89 billion coming from its data center segment, or 92% of its total revenue.
Nvidia's growth and competitive position
Nvidia's revenue grew 106% year-over-year, and data center revenue grew even faster. This dynamic makes Nvidia not only a leader in AI processors, but also a major infrastructure provider. The CNBC report states that Nvidia forecasts revenue growth of 70% in 2028, which would equate to $673 billion in revenue - which would help the company rank second in revenue among U.S. technology companies, behind Amazon.
What do these predictions mean for the technology hierarchy?
Kress says AI labs could surpass traditional tech giants — Apple, Microsoft, Amazon and Alphabet — whose valuations have been fluctuating in recent years. If a single AI lab like OpenAI were to generate $600 billion in computing demand from a single vendor, its overall economic footprint, including revenue from products and services, could be enormous.
It's important to note, however, that these predictions are partly in Nvidia's best interest. Every investor who puts money into the AI labs ends up returning a portion of that money to Nvidia through processor orders. This creates a kind of cycle in which Nvidia and the AI labs reinforce each other.
Conclusions
Nvidia CFO’s prediction that AI labs will become the largest tech companies is based on solid financial data and massive investments in computing infrastructure. If these trends continue, we can expect the tech sector hierarchy to undergo a significant transformation, and AI labs like OpenAI could become not only an engine of innovation, but also an economic leader.
Sources
- Crypto Briefing – Nvidia CFO predicts frontier AI labs like OpenAI will become the largest tech companies in history
- CNBC - Nvidia's 70% growth forecast puts it on track to become tech's No. 2 companies by revenue
- Twit.tv - IM 885: Get on the Butter Box - Can Local AI Models Outperform Frontier Labs?
