The most striking opposition to Amazon's artificial intelligence spending spree is now coming from inside the company. At a Seattle Land Use and Sustainability Committee hearing this month, three Amazon employees testified in favour of tighter rules on data centre construction, speaking publicly against the very infrastructure their employer is racing to build. It is a rare moment of open dissent at a company that has shed roughly 30,000 corporate roles over the past eight months while reaffirming a capital budget of around 200 billion dollars for the year, most of it bound for AI and the data centres that power it.
Workers against the warp speed
The engineers belong to Amazon Employees for Climate Justice, an internal advocacy group whose open letter warned that the all costs justified, warp speed approach to AI development will do staggering damage to democracy, to our jobs, and to the earth. One employee, Patrick Schloesser, told the hearing that Big Tech is desperate to build as much compute capacity as it can, as fast as it can. A colleague, Liesl Wigand, put it more starkly, urging officials not to let Big Tech burn Seattle to win the AI race.
A moratorium, not a memo
The intervention worked. Following public pressure, Seattle City Council voted to impose a yearlong moratorium on new local data centre construction. For once the friction around AI's expansion produced a concrete brake rather than a press release, and it came partly at the urging of the people building the systems. That combination, technical staff lobbying their own city to constrain their employer, captures how fraught the politics of the AI buildout have become.
The capital and the cuts
The backdrop is the uncomfortable arithmetic that has defined 2026. Tech firms have committed hundreds of billions to AI infrastructure while trimming headcount, and Amazon sits near the centre of that story with its 200 billion dollar outlay and tens of thousands of job losses running in parallel. Whether the cuts pay for the compute, or merely coincide with it, the optics are hard to manage when the spending lands as new buildings and the savings land as redundancies.
Why the dissent matters
Internal opposition is easy to dismiss as a fringe, but it changes the conversation. It reframes the AI jobs debate from an abstract argument about automation into a concrete fight over land, power and water in specific cities, with workers as participants rather than bystanders. If more employees follow the Seattle example, the constraint on the AI race may come not only from regulators or markets but from the engineers whose labour the boom depends on.
