Researchers at Northwestern University have made a groundbreaking achievement in artificial intelligence (AI). They developed an AI-based system that can autonomously design robots from scratch.
In a demonstration, the new AI was tasked with creating a robot capable of walking on a flat surface, a feat that would have taken nature billions of years through evolution, yet the algorithm achieved it within seconds.
Fast AI-Driven Design Algorithm
The team noted that what sets this AI apart is its incredible speed and ability to operate on a lightweight personal computer, creating entirely original designs without relying on supercomputers or extensive datasets.
Unlike other AI systems, which often mimic past human works, this AI generates entirely new concepts without being bound by human creativity.
Sam Kriegman, leading the team at Northwestern University, said: "We discovered a very fast AI-driven design algorithm that bypasses the traffic jams of evolution, without falling back on the bias of human designers."
"We told the AI that we wanted a robot that could walk across land. Then we simply pressed a button and presto-it generated a blueprint for a robot in the blink of an eye that looks nothing like any animal that has ever walked the earth. I call this process 'instant evolution'," he added.
Kriegman, an assistant professor of computer science, mechanical engineering and chemical and biological engineering at Northwestern's McCormick School of Engineering, saw this development as a significant leap in exploring artificial life.
The resulting robot, though unassuming in appearance, signifies the first step toward a new era of AI-designed tools that can directly interact with the physical world.
The Future of AI-Designed Robots
The process began with a prompt to design a robot capable of terrestrial locomotion. From there, the AI autonomously iterated through designs, assessed flaws, and refined structures until a functioning walking robot emerged.
Remarkably, this entire process, from a static block to a fully functional robot, took a mere 26 seconds on a standard laptop. Though unconventional, the resulting robot features three legs, fins along its back, and a porous structure.
The researchers believe the porosity adds flexibility and reduces weight, enabling the robot to bend its legs for walking. This achievement opens up a world of possibilities for AI-designed tools.
In the future, robots created by similar programs may navigate hazardous environments, perform maintenance tasks, or even be deployed on medical missions within the human body. Kriegman envisions a future where AI-driven design leads to solutions for some of the most complex challenges faced by humanity.
"When humans design robots, we tend to design them to look like familiar objects," Kriegman said. "But AI can create new possibilities and new paths forward that humans have never even considered. It could help us think and dream differently. And this might help us solve some of the most difficult problems we face."
The team's findings were published in the Proceedings of the National Academy of Sciences.
Related Article: MIT Merges Physics and AI for Image Generation With Poisson Flow Generative Model++
