Nine publicly traded mining companies pulled in $341 million from artificial intelligence and high-performance computing operations during the first half of 2026. Their combined capital spending on those same operations reached $5 billion. That leaves a 15-to-1 gap between what they spent and what they made.
The AI and HPC revenue haul
The $341 million figure represents income generated specifically from AI and HPC workloads, not from the companies' traditional mining activities. It covers the six months ending June 30, 2026. The number is small relative to the scale of investment these firms are making, but it does show that the AI side of their businesses is producing real, if modest, returns.
For context, the average revenue per company works out to roughly $37.9 million over the half. That's not a trivial sum, but it's nowhere near what a $5 billion capital outlay would normally require in short-term payback.
Where the $5 billion went
The capital asset spending of more than $5 billion dwarfs the revenue. This money likely went into data center infrastructure, specialized chips, cooling systems, and other hardware needed to run AI and HPC services. The companies are clearly positioning themselves for a future where these operations generate significant income, but that future isn't here yet.
The spending is a bet. It's a big one. And the companies are making it while the revenue stream is still in its early stages.
A 15-to-1 ratio that raises questions
The ratio of capital expenditure to revenue sits at 15 to 1. That means for every dollar these miners earned from AI and HPC in the first half, they spent fifteen dollars on capital assets. Such a lopsided ratio is unusual for any industry, but it's particularly striking in mining, where margins are already tight.
There's no sign the companies are pulling back. The spending appears to be deliberate, likely aimed at securing a foothold in the AI infrastructure market. But the math is harsh. If the revenue doesn't grow dramatically in the coming quarters, these firms will face serious pressure to justify the outlay to shareholders.
The first half of 2026 is the only reporting period available. Whether the second half brings a surge in AI-related income or another round of heavy spending remains an open question. The companies have not yet provided guidance for the full year.
