For years, commentators compared data to one of the world’s most important commodities: oil. Much like how massive oil reserves conferred wealth and power on the world’s petrostates and fossil-fuel firms in the 20th century, countries and companies with lots of data could leverage that resource in the 21st century.

The argument is taking on greater relevance with the rise of AI: Large countries like the U.S. and China can train their AI models on massive amounts of data, while smaller economies, with less data on hand, just can’t keep up.

But Josephine Teo, Singapore’s minister for digital development and information, argued at the Fortune Brainstorm AI Singapore conference on Tuesday that oil is the wrong comparison to make when talking about how countries use data.

Unlike oil—a finite resource that goes away once extracted and used—data can be “recombined” many times over, she said. “Data is actually more useful when it is reused in many different contexts.”

Singapore adopted an AI strategy in 2019, and is now trying to leverage its existing strengths as a trade and travel hub to bolster its development of the new technology. “We aren’t trying to be an AI superpower. We don’t need to be,” Teo told Fortune in a recent feature on the country’s plans for AI.

On Tuesday, Teo suggested the country’s vibrant economy is one way Singapore can get access to the data it needs. “If you consider the size of our GDP, and use it as an indication of the breadth and depth of activities that are taking place in Singapore, and that every single one of those activities generates a data point—then maybe the data is not as small as we think it is,” she explained.

‘Tropical’ data centers

Still, the Singaporean minister noted some potential challenges to the country’s AI hopes. “In any new field, the pool of talent is always going to be a concern,” she said. “It’s not broad enough, it’s not deep enough.”

But she pushed back against suggestions that Singapore’s brief moratorium on data centers had hindered the country’s attractiveness as an AI hub.

In 2019, the Singaporean government briefly paused approvals of new data centers, owing to concerns about land use and power consumption. Operators fled elsewhere, including to neighboring Malaysia, which is now attracting billions of dollars in investment from Big Tech companies like Microsoft and Google.

But Teo argued Singapore, with 70 data centers, has some of the world’s “densest data center capacity.”

“If you look at our data center capacity relative to the size of our GDP, and if you compare it to, say, Japan or if you compare it to China, actually we have way more,” she claimed.

While Singapore is open to having more data centers, “the question is how do we do it whilst also being able to fulfill our commitments to the net-zero pathway,” she asked.

The answer is “tropical [data center] standards,” she said. “If you’re operating the data center in a tropical climate, do you need to operate it at a temperature that is so low? Or [is it that] actually with one degree higher, you’re okay?”