The Trillion-Dollar Hardware Race and the Invisible Divide
AI-generated image · Bay Street Wire
Big Tech is committing nearly $2.4 trillion to AI infrastructure, but the rush for computing power ignores the human cost of the data that fuels it.
*Opinion: This is a column by Farah Nasser.*
When we talk about the "AI boom," the conversation usually centers on the sheer scale of the capital. According to reporting from Bloomberg News via the Financial Post, the four largest players in the data center race—Alphabet Inc., Meta Platforms Inc., Microsoft Corp., and Amazon.com Inc.—have committed nearly $2.4 trillion in spending for AI infrastructure over the coming years.
To the C-suite, these numbers represent a strategic land grab. Amazon.com Inc. CEO Andy Jassy has compared the current moment to an accelerated version of the initial Amazon Web Services build-out, arguing that massive early spending will yield great returns later. Indeed, AWS reported a 37% revenue jump in the second quarter, the fastest growth since 2021. Even with $220 billion in capital expenditures this year, Jassy claims it is not enough to meet cloud infrastructure demand.
But as a tech columnist focused on access and inclusion, I find the math of this "boom" unsettling. While these companies race to build fleets of data centers, they are fundamentally ignoring the digital divide. The hardware race is designed to capture "massive demand for AI computing power," yet the benefits of this infrastructure rarely trickle down to the marginalized communities whose raw data is the very fuel these systems require to function.
The scale of the commitment is staggering. Alphabet Inc. recently disclosed $902 billion in purchase commitments, contractual obligations, and leases that have not yet commenced—a figure more than nine times higher than it was a year ago. Meta Platforms Inc. has reported nearly $700 billion in future spending, with about half dedicated to data center leases that could be paid off over 30 years.
We are witnessing a historic concentration of power. Alphabet and Amazon have already tipped into negative free cash flow due to this spending, and Meta is expected to follow. This is a gamble on the future of intelligence, but it is a gamble taken by a very small circle of corporate giants.
When Big Tech spends trillions on energy, technical equipment, and server farms, they aren't investing in universal access. They are investing in the walls of their own gardens. The tragedy of the AI boom is that while the infrastructure is being built at a breakneck pace, the people providing the data that makes these models possible remain on the wrong side of the digital divide, excluded from the wealth and the tools that their own information helped create.