The capabilities of large AI systems are constantly improving, but they consume a great deal of energy during training and operation. The human brain, by contrast, is extremely energy-efficient: It requires only around 20 watts.

Researchers at Graz University of Technology, in collaboration with international partners, have developed a novel brain-inspired AI model that can plan flexibly and solve complex problems. In doing so, it consumes significantly less energy than multilayer neural networks or large language models. The study is published in the journal Nature Machine Intelligence.

“The brain works in a completely different way from tod...

In 2010, scientists waiting to use the William Herschel Telescope in the Canary Islands captured a brief 36 minutes of images of Venus. They were waiting for the sun to set so they could look at distant stars and dust disks around young stars...

Wearable health monitors are designed to move with the body. But for many devices, movement creates a problem: The more a person bends, stretches, reaches or runs, the harder it can be for the device to keep a stable wireless connection. Penn State researchers and international collaborators set out to solve that problem by developing a soft, stretchable antenna that can keep working even when pulled in different directions...

Black holes represent some of the most extreme environments in the universe. They are the sources of the strongest gravitational fields, allowing us to test Einstein’s theory of general relativity to an extent impossible with small objects. But we are also reaching the limit of what we can learn about one remotely. So, various authors have put forward ideas for how we might eventually send a probe directly to a black hole to observe it up close. One of the most vocal of those authors is Cosimo Bambi of Fudan University in Shanghai—and he recently released a paper, available as a preprint on arXiv, about what it would take to send a gram-sized probe to a nearby black hole.

Unfortuna...