Economics Gain Scale on AI

Every once in a while, a piece of writing reminds you why real journalism matters, the kind that doesn’t just inform but investigates, questions and connects the dots that others miss.

R. Gurumurthy’s ET Prime article (Oct 10 2025) is one such work (Read here: bit.ly/ETPrimeAI10Oct ). It dives deep into how the AI race today is being driven as much by finance as by technology, a theme that echoed what I found earlier in Gary Rivlin’s AI Valley. Gurumurthy’s analysis “OpenAI-Oracle-Nvidia deals: Big AI moves or just balance-sheet games?”, offers a rare lens on how capital, partnerships and scale are reshaping innovation. I have also been following the product side closely (see my comment on Atlas), it’s striking to see the capital story and the product story converge in real time. 

The article unpacks two eye-catching numbers: OpenAI’s reported plan to buy about $300 billion of Oracle’s cloud capacity over five years and NVIDIA’s possible $100 billion investment that ties back to its own GPU sales. One promises capacity, the other capital, both forming a loop where money, infrastructure and ambition feed into each other.

That’s where it gets interesting.

Put these figures next to OpenAI’s reported revenues, still only in the low teens of billions and the scale gap becomes obvious. It’s not just growth; it’s leverage. As investors celebrate scale, analysts see exposure.

What Gurumurthy calls the financialisation of AI isn’t just a clever phrase. It’s the reality where suppliers become financiers, where today’s capacity booking becomes tomorrow’s revenue claim. It accelerates the ecosystem, no doubt, but also builds invisible walls that make it harder for smaller innovators to even get on the track.

From my view, Sam Altman’s strategy is nothing short of masterful. He’s managed to bind the cloud, chip and platform layers into one cohesive power block. It’s smart, decisive and perfectly timed. But with that elegance come trade-offs like higher entry barriers, deeper dependencies and the risk of markets chasing momentum faster than substance.

Gurumurthy’s reporting does what good journalism should: it lets readers see both sides, the brilliance of coordination and the quiet unease of concentration.

And as we now see with the launch of Atlas, the story is already evolving from financial scale to product ambition. OpenAI seems to be moving seamlessly from building partnerships to owning the user interface, a reminder that in this race, strategy plays out not in isolation, but in layers.

The AI boom isn’t slowing down anytime soon, but as balance sheets become the new algorithms, it’s worth asking, how much of this race is about innovation and how much about valuation?