The study of metallurgy dates back thousands of years, from making tools during the Iron Age to its role in building skyscrapers in the modern era. Now, a new laboratory at Texas A&M University wants to bring twenty-first-century technology to the historic craft.
The College Station-based university announced a new Aggie AI Foundry, a laboratory created by A&M professors Raymundo Arróyave and Ibrahim Karaman that will include a fully automated system to develop metal alloys using artificial intelligence and robotic workers. The university said this would increase the rate at which metal alloys are developed from months to just days.
“I would say that we have a grand vision,” Arróyave told Houston Public Media. “We want to democratize access, accelerate development, and liberate the brains and minds of our students and postdocs so then they use their creativity to advance our field.”
The Aggie AI Foundry is supported by the WoodNext Foundation, a philanthropic organization founded by Roku CEO Anthony Wood and his wife Rusan, both of whom are university alumni. The WoodNext Foundation did not return a request for comment.
Robots will mix, melt, and test materials using computer-controlled arms along an assembly line, according to the university. AI could ultimately test as many as a trillion different combinations of periodic elements.
However, the move has not been without pushback, Arróyave said. Students, researchers, and others have raised concerns with him that a laboratory relying on artificial intelligence and robots rather than humans would dilute the profession of metallurgy.
“I may sound like one of those tech bros that minimize the impact of their frontier models on the workplace,” he said. “I always give my personal experience. When I was a grad student at MIT, I learned how to polish samples. And at the end of a shift polishing samples, I ended up with my fingertips all bloody because my fingers were catching the polishing wheel. It was not that fun. I didn’t learn that much. I wish we had an automated polisher back then. … There are these parts that are not that romantic to the work of a researcher. If we have an automated system, that can make it easier for us.”
An AI-powered laboratory, Arróyave argued, is like a skip-the-line fast pass at a theme park or getting to the Grand Canyon without having to drive there.
“What we’re trying to do is to try to get to the cool, fun part faster,” he said. “But that is not going to negate the careful, detail-oriented process of investigation that happens once you find a really cool material.”
Though the lab is automated, students — and humans, in general — will still play a role in the laboratory. Part of the creative vision is to enable students to focus on higher-level research instead of laboratory work. Arróyave called humans “fundamental” to the process.
“What we’re going to train is a new kind of scientist,” he said. “And they have to learn many more things than what we learned when we were students ourselves.”
