• Tech Tech

Scientists build 'alien coral' copper cooler that could make AI data centers far less thirsty

"This innovation could change the entire debate around data centers."

A man explains data center solutions, with text overlays highlighting a university logo and power usage effectiveness.

Photo Credit: TikTok

One of the biggest challenges facing the AI boom is how to efficiently cool the massive, energy-intensive data centers that power these systems. However, a recent video is gaining traction on TikTok after it explained how a copper object that resembles "alien coral" could be part of the solution.  

In an explainer video discussing recent research, creator @gigaqian discussed how engineers might have identified a more efficient way to cool computers.

@gigaqian Engineers at the University of Illinois had a breakthrough that can help change the way we cool data centers! They partnered with Fabric8Labs in San Diego and are pushing this new tech forward. We might be solving the biggest problem with AI data centers in real time! #breakingnews #datacenters #newtech #futuretech #gigaqian ♬ original sound - GigaQian

What happened?

The video discusses how University of Illinois researchers worked with California-based Fabric8Labs to rethink an important cooling part used in high-performance computers called fins.

According to the creator, the team used topology optimization, an engineering technique that uses algorithms to determine the most efficient arrangement of material, to design ultra-efficient fins. The creator said that these new fin designs were unlike anything a human would normally design and resembled "alien coral."

The results after testing this new design were impressive.

According to the video, early models suggest the new tech could lead to up to 32% better cooling. 

Why does it matter?

Cooling is becoming a much bigger challenge as AI hardware grows more powerful and more energy-intensive.

AI is increasingly tied to the energy grid. It can help forecast electricity demand, improve clean-energy management, and speed up scientific research, but it also comes with serious tradeoffs.

Training and running AI systems can require enormous amounts of electricity and water, and rapid data-center growth can put pressure on local grids, strain infrastructure, and potentially contribute to higher energy bills. There are also concerns about misuse, security, and unintended social impacts.

The impact of ultra-efficient cooling systems could be substantial. Data centers collectively consume more electricity than some countries do, so reducing the energy required for cooling could help cut waste, lower costs, and ease strain on the power grid.

"This innovation could change the entire debate around data centers," the creator wrote in the comment section. 

What are people saying?

Commenters were intrigued, though not entirely convinced.

Many pointed to the price and availability of copper as a major sticking point to advancing the technology. 

Others were more critical of the AI and data center industries themselves, arguing that even with efficiency improvements, the energy these systems consume outweighs the benefits they provide.

Get TCD's free newsletters for easy tips, smart advice, and a chance to earn $5,000 toward home upgrades. To see more stories like this one, change your Google preferences here.

Cool Divider