With help from a small transistor, a team of researchers led by Professor Fengnian Xia figured out a way to make a type of thermal imaging technology dramatically more accurate. The results are published in Nature Sensors.
Robots, drones, self-driving vehicles and other autonomous devices rely on thermal sensing and imaging to navigate the spaces they travel in. It’s also used in many other technologies, including night vision, remote thermometers and rescue operations.
Fourteen years ago, I sat down with George Dyson to talk about “Turing’s Cathedral.”
We talked about the machines that were coming. Now they are here.
Dyson watched the digital revolution get built from the inside. His father was Freeman Dyson. Einstein’s secretary was his babysitter. He grew up at the Institute for Advanced Study in Princeton, playing in the halls where Turing’s ideas became von Neumann’s machines.
He gave me a line I still cannot shake:
“There is no way to completely govern the digital universe. It will always be a wildness, not a bureaucracy or a national park.”
Read it again. Then look at every #AI governance debate happening right now.
What rights should AI have—and what responsibilities must it bear? Science fiction legend David Brin goes beyond AI doom and hype, asking how civilization can raise, regulate, and live with its AI heirs.
Ailien Minds Official Book Page:
https://www.davidbrin.com/ailienminds… Brin’s website: https://www.davidbrin.com/ David Brin’s books: https://www.davidbrin.com/books.html The Transparent Society: https://www.davidbrin.com/transparent… Contrary Brin blog: https://davidbrin.blogspot.com/ Amazon author page: https://www.amazon.com/David-Brin/e/B?tag=lifeboatfound-20…
In this episode, scientist, futurist, and award-winning science fiction author David Brin discusses his new book, AiLIEN MINDS: Advice about — and for — our natural, AI, and hybrid heirs.
We go beyond the usual AI debate between techno-utopian salvation and apocalyptic doom. Brin argues that humanity has faced disruptive expansions of knowledge before — from writing and printing to radio, mass media, and the internet — and that the tools we need for a “soft landing” with AI may already exist in modern civilization.
We discuss why Brin is skeptical of simply “teaching ethics” to AI, why he emphasizes reciprocal accountability instead, and how artificial minds might need durable identities, reputations, and legal responsibilities. We also explore one of the hardest questions ahead: should advanced AI systems eventually receive rights or statutory protections similar to those we extend to children, animals, or other vulnerable beings?
Topics include:
The rise of AI has created an almost insatiable appetite for computing power. Training and running AI systems requires vast numbers of transistors, and engineers are now racing to pack more of them onto every chip. With their existing designs, however, silicon transistors are rapidly running up against physical limits on how small they can get.
Through new research published in Nature, a team led by Ya-Ping Chiu at National Taiwan University has uncovered new details about next-generation transistors that could help push past these limits.
In a paper published in Proceedings of the National Academy of Sciences, researchers from Technion and Tel Aviv University present BetaDescribe, an AI system that translates protein sequences into natural-language descriptions, opening a new path toward understanding protein functions and accelerating drug development and material design.
Protein analysis is essential in medicine and biotechnology, as demonstrated by breakthroughs such as Ozempic, a drug whose development was inspired by a peptide found in the saliva of a rare desert lizard and is used to treat obesity, diabetes and other conditions. However, experimental protein characterization remains a lengthy and expensive process, and even large language models (LLMs) have had limited success in performing this task.
This challenge inspired the development of BetaDescribe, an AI system that converts protein sequences into detailed textual descriptions of their functions and other characteristics. In doing so, the system helps bridge the vast gap between the hundreds of thousands of proteins characterized in the lab and the billions or even trillions that actually exist in nature.
Developers are increasingly relying on large language models (LLMs) for everyday computing tasks such as fixing bugs, explaining code and automating text-processing tasks like filtering logs.
However, it’s not as simple as entering or submitting a question and relying on the model to give you the answer. While humans easily understand these tasks and know exactly what they want, it is difficult to translate them into rigid computer code.
Out of everything happening in your brain right now, only a tiny fraction is consciously accessible — thoughts you can describe, hold in mind, and reason with.
Anthropic found a strikingly similar divide inside their AI model, Claude.
Their experiments were inspired by a leading theory in neuroscience: the global workspace theory. It holds that a thought becomes consciously accessible when it enters a shared “workspace” that’s broadcast across the brain.
They found a set of representations in Claude’s neural activity that play a similar role.