Isabelle Reksopuro, a student at the University of Washington, has built an interactive map of AI data center policy that pulls from Epoch AI's dataset and scraped legislation, and refreshes itself four times a day using Claude. The map tracks construction, local opposition, and the patchwork of state-level rules now governing where hyperscalers can put compute. It is the kind of civic-tech artifact that exists because the official version doesn't.
The trigger was Reksopuro's home state of Oregon. The Dalles, a city of 16,010 as of the 2020 census, asked to take ownership of a 150-acre portion of Mount Hood National Forest, citing municipal water needs as its population grows. Critics argue the real beneficiary is Google's data center campus in The Dalles, which already consumes about one-third of the city's water supply.
"There's a lot of misinformation about data centers," Reksopuro said. "Google has denied taking that land." Technically, she notes, the city is the entity seeking the land, with Google sitting downstream as "a big, unnamed power user." That opacity is what the map is designed to puncture.
Key facts
- 01Isabelle Reksopuro, a University of Washington student, built an interactive map of global AI data center policy using Epoch AI data and legislation scraping.
- 02The map uses Claude to search for new sources four times a day and auto-populate a news feed.
- 03Google's data center campus in The Dalles, Oregon (population 16,010) consumes about one-third of the city's water supply.
- 04The Dalles asked for ownership of a 150-acre portion of Mount Hood National Forest, citing municipal water needs.
- 05Texas gives data center operators more than $1 billion in tax breaks each year; Maine's hyperscale moratorium passed in April was vetoed by Governor Janet Mills.
On the technical side, Claude does the heavy lifting. "Once it does that, it will write a new summary, add it to the news feed, and populate it on the sidebar," Reksopuro said of the model's role in the pipeline. "I wanted it to be self-updating, since I'm also a student." The map was built to be readable by her younger sisters rather than policy wonks — a deliberate design choice aimed at TikTok-grade attention spans.
“Google's data center campus in The Dalles already consumes about one-third of the city's water supply, while Texas hands data center operators more than $1 billion in tax breaks each year.”— Jaeden Schafer
The picture that emerges is uneven. Maine passed the first state-level moratorium on hyperscale data centers in April, before Governor Janet Mills vetoed it. Texas is on the opposite end: the state passed a tax exemption for data center operators and now hands the sector more than $1 billion in tax breaks each year, per The Texas Tribune.
Opposition to data centers is one of the rare issues that crosses party lines in the United States. After the construction phase, hyperscale sites bring few permanent jobs, and the load is now visibly pushing power costs higher across much of the grid. That tension — local cost, distant benefit — is what's driving the patchwork of moratoria, tax fights, and zoning battles the map is trying to catalogue.
Reksopuro isn't opposed to data centers as a category. Her argument is that information asymmetry favors operators. "Right now, it's this really opaque thing — and all of a sudden, there's a facility," she said. "I think that if people knew about data centers beforehand, it would give them leverage. They would be able to negotiate: ask for job training programs, tax revenue, environmental monitoring, things to improve their community."
The map sits inside a wider conversation AI Chat Daily has covered recently, including Hugging Face's Sasha Luccioni launching the Sustainable AI Group to push standardized reporting on AI infrastructure impact. The throughline is the same: data on data centers is fragmented, and the people building the dashboards are often researchers and students rather than the operators themselves.
The obvious caveat is that a scraped-and-summarized policy feed is only as accurate as its inputs. Claude summarizing local legislation four times a day will miss nuance that a municipal attorney wouldn't, and the map can't independently verify operator-water or operator-power claims that companies dispute. Reksopuro acknowledges the misinformation problem directly — it's part of why she built the thing.
For the AI industry, the signal is that the siting fight is now legible to the public in a way it wasn't a year ago. Hyperscalers have built a business model that depends on quiet permitting, cheap water, and tax abatements negotiated before residents know a facility is coming. A self-updating map maintained by a college student doesn't change the economics, but it shortens the window between proposal and organized response — and that window is where the $1 billion-a-year Texas-style deals get done.
Working on something we should cover, or seeing a story we missed? Send leads, documents, or feedback to hello@aichatdaily.com. For sensitive tips, see our secure tips page for Signal and PGP options.
Spotted an error? Email hello@aichatdaily.com with the URL and the issue, or read our full corrections policy.
