Contrary to the narrative of burning cash, major AI labs are likely highly profitable on the marginal cost of inference. Their massive reported losses stem from huge capital expenditures on training runs and R&D. This financial structure is more akin to an industrial manufacturer than a traditional software company, with high upfront costs and profitable unit economics.

An NVIDIA director highlights a significant, under-the-radar growth vector: accelerating traditional enterprise software. Oracle's decision to run its classic database on GPUs represents a trillion-dollar infrastructure shift from CPUs to GPUs for core business applications, proving NVIDIA's market extends far beyond the current AI boom.

Despite Microsoft's massive AI investments, its stock only grew 4%, while NVIDIA's market cap soared. Investors punished Microsoft's heavy capital expenditure, favoring NVIDIA’s high-margin, fabless "picks and shovels" approach that captured immediate AI profits without the same infrastructure risk.

Oracle is mitigating the immense capital expenditure of its AI cloud buildout by allowing customers to provide their own hardware. This 'BYOH' model, while still a small part of its business, reassures investors by allowing Oracle to expand capacity without footing the entire bill for expensive GPUs.

The huge CapEx required for GPUs is fundamentally changing the business model of tech hyperscalers like Google and Meta. For the first time, they are becoming capital-intensive businesses, with spending that can outstrip operating cash flow. This shifts their financial profile from high-margin software to one more closely resembling industrial manufacturing.

Oracle's significant investment in AI infrastructure appears less risky because they've structured deals where major clients like Meta and OpenAI pay for GPUs upfront or bring their own hardware. This strategy prevents Oracle from becoming overleveraged while rapidly scaling its data center capacity.

To finance its capital-intensive AI cloud build-out for customers like OpenAI, Oracle may create the first public "chip-backed asset-backed security" (ABS). This novel financial instrument would let Oracle raise money against its existing GPUs in public markets, lowering costs and potentially keeping debt off its balance sheet via a special-purpose vehicle.

Oracle's stock is trading near the value of its remaining performance obligations ($523B RPO vs. $568B market cap). This suggests investors are heavily discounting the future profitability of its massive AI data center deals, questioning the long-term economics of being a commodity compute provider.

Traditional SaaS metrics like 80%+ gross margins are misleading for AI companies. High inference costs lower margins, but if the absolute gross profit per customer is multiples higher than a SaaS equivalent, it's a superior business. The focus should shift from margin percentages to absolute gross profit dollars and multiples.