Moe Asnani has been in the weed business for over 13 years. He’s the former owner of two D2 Dispensaries here in Tucson. He even started his own cannabis brand, iLAVA, to make THC-based pain-relief gels.
But recently, he’s set his sights on a new project — converting old cannabis farms into mini data centers.
“For some reason, there's a lot more cannabis grown in Arizona than Arizonans use,” Asnani said. “And so a lot of folks are getting out of that business, like myself, at least on the cultivation side. But these facilities can potentially be repurposed to be mini data center nodes.”
Mini data center nodes. What does that mean exactly?
To understand that, you need to first understand what a data center is. According to David Ebert, Chief AI and Data Science Officer at the University of Arizona, at their most simple level, data centers are “rooms or facilities where computing is done.”
Data centers can be as large as the controversial Project Blue facility going up near the Tucson fairgrounds, or as small as a rack of servers on a desk – something Asnani himself has.
What Asnani is proposing lies somewhere in the middle — something large enough to serve a couple hundred thousand users, but small enough that its water and power footprint is roughly that of an indoor cannabis growth facility.
Asnani is correct that the cannabis industry is in decline. In Arizona, cannabis sales peaked in 2024 and have been declining ever since, a pattern experts believe is due to an oversupply driving down prices.
Because indoor cannabis farms already require relatively large amounts of power and water, Asnani says they are well suited to being retrofitted as mini data centers.
The concept is called “adaptive reuse”, and it has the attention of Tucson City Council member Paul Cunningham.
“Data centers are dirty words and everybody knows that. And no one wants to build a brand new data center anywhere,” Cunningham said. “But if we have buildings that were using that amount of water and that amount of energy already in their previous function and [it] can be done without being invasive, I'm OK with exploring that option.”
"Data centers are dirty words and everybody knows that."Paul Cunningham, Tucson City Council
Because Arizona does not require growers to report their water usage, it’s hard to find exact numbers for how water-intensive cannabis farming is in the state.
Asnani’s former company, D2, estimated that its 18,000-square-foot growing facility used about 1.7 million gallons of water in 2024. For comparison, the controversial Project Blue data center being built in Tucson could use up to 31 million gallons per year.
Beyond their lower levels of water and power usage, Asnani says these mini data centers would allow for what he calls “sovereign AI” — a system in which users could control all aspects of their data processing rather than outsourcing to hyperscale data center operators.
He also claims that these small data centers would make users’ data more secure from cybersecurity threats.
“The whole idea is if you create these smaller nodes where people control their own data, where they're not at the mercy of other people's updates and things like that – that's kind of where the idea came in, because I believe in decentralization,” Asnani said.
Asnani isn’t the only one looking to build smaller data centers. OpenAI and Anthropic, two of the country’s leading AI companies, have recently invested in a slew of small data centers, some on the scale Asnani is proposing.
Differing opinions
According to Ebert, while the idea of smaller scale data centers might be appealing, they could actually be less efficient than larger-scale projects.
“The fact of thinking of the large data centers as being less efficient in terms of environmental impact really isn't well proven,” he said. “Most large scale data centers have a power use efficiency just above 1.0, where most small scale data centers have it closer to 2.0 or 2.8. And the higher the number, the less efficient it is.”
That’s partly because those hyperscale projects share the load among thousands of users, making sure that every bit of power is used productively. Small data centers used by fewer clients, by contrast, could sit idle more often.
Ebert also pushed back on the idea that smaller data centers are more secure, saying that network safeguards and professional cybersecurity staff are more important than the number of users or connection points.
"You're using AI all the time, whether you actively turn it on or not. And it is in things you rely on every day."David Ebert, Chief AI and Data Science Officer at the University of Arizona
“If you think of all of your credit card companies, your banks… they all use large scale data centers,” he added.
Still, he says that there may well be a market for Asnani’s idea.
“Just because nothing comes to my mind where having a localized one where maybe it's used by ten companies has an advantage, doesn't mean that there isn't one,” Ebert said. “And it's what an entrepreneur and innovator can do and create that demand and show its advantages. So I think everything's worth exploring.”
Betting on chips
Both Ebert and Asnani expect microchips and computers to get much more efficient in the next five years — a belief backed up by experts — but they disagree on what that means for data centers.
Asnani predicts that more efficient chips will undercut the case for hyperscale data centers because smaller facilities will be exponentially more powerful than they are today.
“If the same laws apply in terms of evolution of microchips and computers, there’ll be something more power efficient by 2029 or 2030,” he said.
Asnani believes there will be less of a need for massive projects when a smaller one can do the job.
Ebert agrees that chips will be many times more efficient by 2030. But he thinks that evolution will hurt small centers more, as they will need to constantly upgrade their hardware in order to follow the pace of evolution. Also, he says, the upfront cost needed to build efficient cooling systems may be prohibitive for a smaller operation.
“Technology that doesn’t use water or uses much less electricity or cooling is something that could be a higher cost at the beginning,” Ebert said. “As it grows and the economies of scale take off, it becomes more efficient.”
"It’s an easy transition to say, ‘We'll use this building for something else.’"Moe Asnani, Tucson entrepreneur
Data centers, big or small, are here to stay
What Cunningham, Ebert and Asnani do agree on is the fact that data centers, whether people like them or not, aren’t going anywhere.
“You're using AI all the time, whether you actively turn it on or not. And it is in things you rely on every day,” said Ebert.
Councilman Cunningham says his support of adaptive reuse is not an endorsement of AI or data centers, but a choice to reduce inevitable impact.
“I think the public opposition will be there no matter what, no matter what the data centers try to do,” he said. “But what I'll tell you is that repurposing buildings and putting restrictions on their water and energy use is way better than building brand new monstrosities on the outskirts that are totally invasive to our way of life.”
As of now, Asnani is only in the initial stages of putting his idea into motion. He says his first step is to convince local government and the public of the utility of his idea.
“It’s an easy transition to say, ‘We'll use this building for something else.’ The government right now makes it pretty complex to go through that process,” he said. “I think making that easier and saying, ‘OK, is your power usage going to be more than before? Is your water usage going to be more than before?’ If those things are easy to answer and have the right answer, then I think it shouldn't be hard.”
