We present SensorFM, a foundation model for wearable health pre-trained on more than one trillion minutes of sensor data from five million people. By co-scaling model size and data, SensorFM learns a general-purpose representation of human physiology that transfers to 35 health prediction tasks, supports label-efficient adaptation and data infilling, and can serve as a grounding tool for a Personal Health Agent.
Multi-platinum recording artist Sia closed the Breakthrough Prize ceremony with an inspiring rendition of “Unstoppable” as all prize laureates returned to the stage to a standing ovation.
__
The eleventh Breakthrough Prize awards celebrated outstanding scientific achievements, honoring scientists driving remarkable discoveries in gene editing, human diseases, the search for the fundamental laws of the Universe and pure mathematics. Held at the Barker Barker Hangar in Santa Monica, CA, presentations were given by Christina Aguilera, Drew Barrymore, MrBeast, Lily Collins, Vin Diesel, Jodie Foster, Gal Gadot, Salma Hayek Pinault, Ke Huy Quan, Gayle King, Edward Norton, Gwyneth Paltrow, Seth Rogen, Lauren Sanchez, Jeremy Strong, will.i.am, and more. With live performances by Katy Perry and Sia. Continued at https://breakthroughprize.org/News/92.
Full show: • 2025 Breakthrough Prize Ceremony: Full Show.
For the first time, physicists have built a working version of quantum mechanics without complex numbers — numbers that have been considered essential to the theory for nearly a century.
Complex numbers combine a regular “real” number with an “imaginary” one — a multiple of the square root of-1, represented by the symbol i — into a single value, like 3 + 4i. The square root of-1 doesn’t correspond to any quantity you could count or measure directly (you can’t have negative one apple, for instance), which is why mathematicians call it imaginary.
Not exactly—but how we use it matters.
A new Trends in Cognitive Sciences perspective argues that AI doesn’t inherently erode human intelligence. Instead, it highlights a well-known principle in cognitive psychology: cognitive offloading.
When we let AI perform tasks that require reasoning, writing, memory, or problem-solving, we reduce the amount of mental practice our brains receive. Like physical exercise, cognitive skills strengthen through use and weaken through disuse.
Skills: learned abilities such as writing, mathematical reasoning, diagnosis, or programming. These are most vulnerable if AI consistently replaces the learning process.
Basic cognitive abilities: foundational functions like working memory, attention, and executive control. Current evidence suggests these may be more resistant to decline, although more research is needed.
The key message isn’t that AI makes people “stupid.” Rather: AI can improve immediate performance. Overreliance may reduce long-term learning and skill retention.
AI is most beneficial when it augments human thinking instead of replacing it. This fits with decades of neuroscience showing that practice drives neuroplasticity. The brain adapts to the cognitive demands we place on it. If.
If you were able to create an immortal version of yourself, would you? Until this decade, that question was the stuff of science fiction, but now experts in the fields of artificial intelligence and robotics suggest it will indeed be possible. This cinematic documentary explores the latest technological advancements in AI, robotics and biotech, and poses the question: what is the essence of the human mind, and can this be replicated? Or even more unsettling, could we one day meet cloned versions of ourselves – clones which are better, smarter, and immortal?
Stars: Bina48, Nick Bostrom, Lincoln Cannon.
This is under license from Sideways. All rights reserved.
Welcome to Machina! Where you can find everything and anything about cars, motorbikes, planes, trains and much more!
SUBSCRIBE FOR MORE MACHINA: https://bit.ly/3VgYf2o.
Do not forget to click the notification bell to get more Machina videos like this and share your experience.
Your support fuels our exploration! Help us uncover new stories and share diverse voices by donating to our channel through Super Thanks!
In green are VTA neurons receiving inputs from the lateral habenula, in red are VTA neurons projecting back to the lateral habenula, and in blue are dopaminergic (TH+) VTA neurons. Approximately 25% of VTA neurons exhibit both presynaptic and postsynaptic connectivity with the lateral habenula. Lateral habenula neurons projecting to the VTA encode unpleasant signals that are necessary for learning to avoid or respond to threats.
🔗 Use the link in our bio to see the article by Marina R. Ihidoype et al. in the July 8, 2026, issue of #JNeurosci for more information. ㅤ 📸 Cover image: Marina Ihidoype.
This week we’re talking about photonics. My guests are Tim McKenna and Ryo Yanagimoto from the Physics and Informatics Laboratories at NTT.
Tim and I chat about balancing theoretical physics with real-world applications at NTT, his most exciting photonics projects, and the primary obstacles to replacing traditional electronics with photonics technologies.
Ryo and I dig into the game-changing potential of NTT’s programmable photonics chips. We discuss how their unique reconfigurability is shaking up traditional hardware manufacturing, facilitating a move from power-hungry electrical processing toward light-driven computation, even allowing chips to self-correct for environmental shifts.
“We can’t win against technology. We are technology.”
Daniel H. Wilson said that to me in 2012. Robotics PhD out of Carnegie Mellon, New York Times bestselling novelist, the guy Spielberg optioned for Robopocalypse. Back then, the line landed like a sharp bit of science fiction.
Fourteen years later, it reads less like a provocation and more like a diagnosis.
His novel Amped was about what happens when technology stops being a tool you hold and becomes part of the body you are. In 2012, that was speculative. Now there are chips being implanted in human skulls, and companies are racing to sell you cognitive upgrades. The “superhuman” future Daniel described is being built now, while most people are still debating whether it will show up at all.
What stuck with me most was that he refused the tidy doom story. He didn’t buy that a superhuman AI would spend its existence trying to exterminate us. That’s a human fear projected onto something that owes us nothing. The harder question, the one worth sitting with, is what we become when the enhancement is not a gadget in our hand but us.
Pulled from the archive and worth another look. One of 300+ conversations on #SingularityFM about where #AI and human #enhancement actually lead, not where the marketing promises they will.
Geoscientists at the University of Glasgow have helped reveal new evidence about the formation of one of the highest mountainous areas on Earth—the Tibetan Plateau. A study by an international team of Chinese and U.K. geoscientists shows that the unique topography at the summit of the plateau is shaped by processes going on deep in Earth.
These features clearly indicate how far the Indian continental plate, to the south, has been pushed beneath the Asian plate, to the north, highlighting the connection between Earth’s interior and its surface features.
Using geochronological analyses pioneered at the University of Glasgow and the Scottish Universities Environmental Research Center, the study team determined that the western and central parts of the Tibetan Plateau have distinct geological histories, reflected in their topography.