Google Burned More Cash Than It Made Last Quarter, Thanks to AI

Google just posted one of the strangest quarters in its history: record-breaking revenue and a negative bank balance, at least by one key measure. The company reported $119.8 billion in total revenue for the second quarter of 2026, comfortably beating what analysts expected. And yet its stock dropped anyway, because Google also revealed it's now spending more on AI infrastructure than it's bringing in.

The culprit is capital expenditure, or capex, the money companies pour into physical infrastructure like data centers, chips, and servers. Google spent $44.9 billion expanding its AI footprint in Q2 alone. Its operating cash flow for the quarter, after stripping out non-cash investment earnings, came to about $39.1 billion. Subtract one from the other and Google ended the quarter with -$5.8 billion in free cash flow, the first time that number has gone negative in the company's history, according to Ars Technica.

That's a notable milestone for a company that has spent two decades as one of the most reliably profitable businesses on the planet. It's not that Google is suddenly struggling. Its core business is thriving. It's that the company has decided to spend even more aggressively than its already enormous income allows, betting that the AI buildout will pay off later.

Where the money came from

Search remains Google's biggest earner by far, pulling in $63.3 billion for the quarter. But the more interesting number is Google Cloud, which brought in $24.8 billion, a 23.8 percent jump from the previous quarter. That kind of growth points to serious demand for Google's AI infrastructure and services from businesses building their own AI products on top of Google's platform.

The company's subscriptions, platforms, and devices business added another $12.9 billion, while YouTube ads brought in $11.1 billion, up more than 12 percent quarter over quarter. Notably, Google grew that YouTube number partly by making ads longer, a small but telling detail about where the company is finding extra revenue as it hunts for cash to fund its AI ambitions.

Once you back out non-operating investment gains, Google's actual operating cash flow of $39.1 billion represents a healthy 40 percent increase over the same quarter last year. In a normal environment, that would be the headline. Instead, it's just the number that got dwarfed by what Google spent building AI capacity.

The spending is only going up

The scale of Google's AI investment becomes clearer when you look at the trajectory. The company spent $91 billion on infrastructure in all of 2025. Heading into this year, Google had told investors to expect capital expenditures between $180 billion and $190 billion for 2026, already double the prior year's total. Now the company says that figure could climb as high as $205 billion.

That's an eye-watering jump in the span of a single year, and it puts Google squarely in league with the other major tech companies pouring unprecedented sums into AI data centers, chips, and power infrastructure. The logic behind the spending is straightforward: training and running large AI models requires massive computing capacity, and companies that fall behind on that infrastructure risk falling behind on the AI products built on top of it. Google Cloud's growth suggests that bet is at least partly working, since customers are clearly paying to use the infrastructure Google has built.

But the negative free cash flow quarter is a signal that the arms race has real financial consequences, even for a company as profitable as Google. When your spending on future capacity outpaces your current cash generation, something has to give eventually, whether that's slower spending, higher prices, or a bet that revenue from AI products will eventually catch up to the cost of building them.

For now, investors seem to be watching that math closely. Despite Google beating revenue expectations across nearly every part of its business, the stock fell on the news, a reaction that suggests Wall Street is paying as much attention to how much Google is spending on AI as it is to how much money the company is actually making from it. The next few quarters will show whether that spending starts translating into revenue growth big enough to justify it, or whether Google, like its rivals, is simply in the expensive early innings of a much longer AI buildout.