Senator Markey Proposes Federal Certification for AI Datacenters
Massachusetts Democrat unveils sweeping legislative package targeting datacenter environmental impact, automated hiring systems, and child safety risks.

Senator Ed Markey is preparing to introduce legislation that would require federal certification for datacenters before construction begins, marking one of the most ambitious regulatory proposals yet for the AI infrastructure boom.
The Massachusetts Democrat unveiled a comprehensive "AI accountability agenda" on Friday that includes nearly a dozen bills targeting what he describes as the unchecked expansion of artificial intelligence across American society. The datacenter certification proposal, which Markey plans to formally introduce in coming weeks, represents the centerpiece of his regulatory push.
Federal oversight for datacenter construction
According to a preliminary draft shared with The Guardian, Markey's datacenter bill would require companies to obtain certification from the Federal Communications Commission before building new facilities. The FCC would evaluate potential impacts on air and water quality, noise levels, energy costs, electricity reliability, local ecosystems, and economic effects on surrounding communities.
The legislation mandates consultation with federal, state, and local agencies including the Environmental Protection Agency and zoning boards. Markey framed the proposal as a response to mounting concerns about datacenters' environmental footprint, stating: "We need to make sure these datacenters don't turn into pollution bombs."
Broader AI accountability measures
The datacenter proposal sits alongside several other bills in Markey's package. Key provisions include:
- Banning employers from relying primarily on automated systems for hiring, firing, and promotion decisions
- Requiring AI chatbot companies to implement safeguards preventing emotional dependency in children
- Mandating independent audits for bias and discrimination before releasing decision-making algorithms
- Creating civil rights offices within every federal agency using or overseeing AI
- Establishing human override options for AI decisions in healthcare facilities
- Protecting workers who disagree with AI recommendations
Markey cited specific cases to illustrate the human cost of unregulated AI, including parents whose 14-year-old son died by suicide after interactions with a chatbot, a Georgia resident unable to drink tap water after datacenter construction, and a nurse experiencing moral distress from following AI recommendations over professional judgment.
Legislative prospects
Despite minimal federal action on AI regulation since ChatGPT's 2022 release, Markey expressed optimism about eventual bipartisan support. One of his proposals has already gained traction: the Senate passed the Children and Teens' Online Privacy Protection Act in March, which would ban targeted advertising to minors and restrict data collection.
The 79-year-old senator, running for his third full term, argued that state-by-state approaches leave too many people unprotected. "Every American is entitled to these safeguards," he said, emphasizing the need for national standards rather than piecemeal regulation.
Why it matters
Markey's legislative package represents the most detailed federal attempt yet to address AI's infrastructure and societal costs simultaneously. While previous proposals have focused narrowly on algorithmic bias or child safety, this agenda connects datacenter environmental impact with workplace automation and economic inequality. The datacenter certification requirement, if enacted, would establish unprecedented federal oversight of AI infrastructure before facilities are built—a significant departure from the current reactive regulatory approach. However, the bills face an uphill battle in a Congress that has shown little appetite for comprehensive tech regulation.
These details were first reported by The Guardian in an exclusive interview with Senator Markey.
This is an original analysis by the Omega editorial team. Source reporting: AI Watch.
Want systems like this working for your business?
Book a Call
