OpenAI’s Hunger Grows, So Do Bills

Artificial intelligence is everywhere right now — writing emails, coding software, answering questions, and quietly reshaping how businesses operate.

At the center of this boom sits OpenAI, the company behind ChatGPT, widely seen as both the symbol and the engine of the generative AI revolution.

But behind the glossy demos and trillion-dollar valuation chatter is a far less glamorous reality: AI is ravenously expensive. Every smarter model demands more chips, more data centers, more electricity — and far more capital than traditional tech businesses ever needed.

As OpenAI’s influence grows, so does a critical question haunting investors and markets alike: Can the economics of AI ever catch up with its ambition?

This is no longer just an OpenAI story. It’s a test case for whether the entire AI boom is building toward sustainable profits — or inflating the most capital-intensive tech bubble in history.

OpenAI Is Getting Better At Money — Just Not Enough

There is real progress. OpenAI has significantly improved its compute margins, meaning it now keeps more revenue after paying the massive costs of running its models.

Margins reportedly jumped to 68% by late 2025, a sharp improvement from just a year earlier and better than some rivals on a per-user basis.

That’s a big deal. It shows pricing power, enterprise demand, and operational improvements.

But margins aren’t profits — and OpenAI is still very much in the red.

The Trillion-Dollar Appetite Behind Intelligence

Here’s where the story turns uncomfortable.

OpenAI has committed — or is expected to commit — up to $1.4 trillion to AI infrastructure over the next decade. That includes sprawling data centers, long-term cloud contracts, and enough electricity to power something close to a U.S. state.

Banks like HSBC estimate that even with explosive user growth and more than $200 billion in annual revenue by 2030, OpenAI would still face a $200+ billion funding gap. In plain terms: the company may get bigger, more influential, and more efficient — and still not make money.

One analyst described OpenAI as “a money pit with a website on top.” It’s harsh, but it captures the scale of the challenge.

IPO Dreams Meet Data-Center Reality

This backdrop explains why IPO speculation won’t go away.

OpenAI says going public isn’t its focus — yet CEO Sam Altman has repeatedly acknowledged that an IPO is the most likely long-term path.

Reports suggest OpenAI could eventually seek a valuation as high as $1 trillion, making it one of the largest IPOs ever.

But markets are asking tough questions: Is this a software company with margins that will expand over time — or a utility-like business permanently tied to massive capital spending?

The answer matters not just for OpenAI, but for the entire AI trade.

The AI Bubble Question No One Can Ignore

None of this means AI is fake, useless, or going away. Demand is real. Productivity gains may still come. OpenAI’s technology is undeniably powerful.

But history offers a warning: revolutionary technologies often arrive long before their economics make sense. Railroads, electricity, the internet — all created enormous value, but not without bubbles, crashes, and painful resets along the way.

OpenAI is now the clearest embodiment of that tension. It’s building the future — one enormously expensive data center at a time.

The question markets are quietly asking is simple and unsettling: How long will investors keep feeding AI’s hunger before demanding results?

That's all for this week! If you found these updates useful, you'll like more from this newsletter. Get deeper dives, hot takes, and all the latest tech news delivered straight to your inbox.