Lifeboat Foundation: Safeguarding Humanity
Toggle light / dark theme

Blog – Page 10

Get the latest international news and world events from around the world.

Log in for authorized contributors

AI model offers accurate and explainable insights to support autism assessment

Scientists have developed and tested a deep-learning model that could support clinicians by providing accurate results and clear, explainable insights—including a model-estimated probability score for autism.

The model, outlined in a study published in eClinicalMedicine, was used to analyze resting-state fMRI data—a non-invasive method that indirectly reflects via blood-oxygenation changes.

In doing so, the model achieved up to 98% cross-validated accuracy for Autism Spectrum Disorder (ASD) and neurotypical classification and produced clear, explainable maps of the brain regions most influential to its decisions.

Shape-shifting material could transform future of implantable and ingestible medical devices

Researchers led by Rice University’s Yong Lin Kong have developed a soft but strong metamaterial that can be controlled remotely to rapidly transform its size and shape.

The invention, published in Science Advances, represents a significant advancement that can potentially transform ingestible and .

Metamaterials are synthetic constructs that exhibit unusual properties not typically found in . Instead of relying solely on , the effective behavior of these materials is primarily determined by the physical structure, i.e., the specific shape, arrangement and scale of their building blocks.

‘Rhythm beats volume’: How the brain keeps the world looking familiar

The brain is famously plastic: Neurons’ ability to change their behavior in response to new stimuli is what makes learning possible. And even neurons’ response to the same stimuli changes over time—a phenomenon known as representational drift. Yet our day-to-day perception of the world is relatively stable. How so?

Resolving such puzzles matters for future brain-computer interfaces, sensory prostheses and therapies for neurological disease. On a quest for an answer, Rice University scientists have built ultraflexible probes thousands of times thinner than a and used them to track neurons in the visual cortex of mice for 15 consecutive days as the animals viewed thousands of images—from line patterns to pictures of the natural world.

The devices, called nanoelectronic threads (NETs), embed seamlessly with , allowing for high-fidelity chronic recordings of .

Chemists create light-switchable magnets that remain active for hours

A research team from the University of Chemistry and Technology, Prague (UCT Prague) and the Institute of Organic Chemistry and Biochemistry of the Czech Academy of Sciences (IOCB Prague) has created and described a new type of photoswitch. The molecule, a thienyl-based acylhydrazone, undergoes an unprecedented “closed-to-open-shell” transformation, where light converts it into a stable diradical.

While previously published lifetimes of such triplet states are a few milliseconds, this ’s switched state has a half-life of over six hours. This revolutionary innovation opens the way for optimizing , developing new and spintronic devices, and targeted elimination of antibiotic-resistant pathogens. The work is published in the Journal of Materials Chemistry C.

Photoswitches are molecules that change between two states under the influence of light. This new switch is unique because it transitions from a stable, non-magnetic (closed-shell) state to an exceptionally long-lived magnetic (open-shell triplet) state. In this triplet state, two electrons have parallel spins, making the molecule paramagnetic and highly reactive. This state is crucial for many photochemical processes, including the generation of .

Primordial black hole’s final burst may solve neutrino mystery

The last gasp of a primordial black hole may be the source of the highest-energy “ghost particle” detected to date, a new MIT study proposes.

In a paper appearing today in Physical Review Letters, MIT physicists put forth a strong theoretical case that a recently observed, highly energetic neutrino may have been the product of a primordial black hole exploding outside our solar system.

Neutrinos are sometimes referred to as ghost particles, for their invisible yet pervasive nature: They are the most abundant particle type in the universe, yet they leave barely a trace. Scientists recently identified signs of a neutrino with the highest energy ever recorded, but the source of such an unusually powerful particle has yet to be confirmed.

Novel catalyst design could make green hydrogen production more efficient and durable

A new type of catalyst—a material that speeds up chemical reactions—that could make the production of clean hydrogen fuel more efficient and long-lasting has been developed by a team led by City University of Hong Kong, including researchers from Hong Kong, mainland China, and Japan.

This breakthrough uses high-density single atoms of iridium (a rare metal) to greatly improve the process of splitting water into and , which is key to like hydrogen fuel cells and large-scale energy storage.

The researchers created a highly stable and active by placing single iridium atoms on ultra-thin sheets made of cobalt and cerium compounds. Called CoCe–O–IrSA, the final product performs exceptionally well in the water-splitting process. It requires very little extra energy (just 187 mV of overpotential at 100 mA cm-2) to drive the oxygen evolution reaction at a high rate, and it stays stable for more than 1,000 hours under demanding conditions.

New light-powered gears fit inside a strand of hair

Researchers at the University of Gothenburg have made light-powered gears on a micrometer scale. This paves the way for the smallest on-chip motors in history, which can fit inside a strand of hair. The research is published in the journal Nature Communications.

Gears are everywhere—from clocks and cars to robots and wind turbines. For more than 30 years, researchers have been trying to create even smaller gears in order to construct micro-engines. But progress stalled at 0.1 millimeters, as it was not possible to build the drive trains needed to make them move any smaller.

Researchers from Gothenburg University, among others, have now broken through this barrier by ditching traditional mechanical drive trains and instead using to set the gears in motion directly.

MicroBooNE detector excludes electron neutrino cause of MiniBooNE anomaly

A recent Physical Review Letters publication presents a thorough analysis of MicroBooNE detector data, investigating the anomalous surplus of neutrino-like events detected by the preceding MiniBooNE experiment.

In 1990, the LSND (Liquid Scintillator Neutrino Detector) experiment observed an anomalous signal indicating the potential existence of sterile neutrinos—a fourth neutrino species beyond the three established flavors (electron, muon, and tau neutrinos).

MiniBooNE was constructed to examine this anomaly utilizing the same neutrino beam methodology. However, instead of resolving the mystery, MiniBooNE discovered an anomaly of its own.

‘Like talking on the telephone’: Quantum computing engineers get atoms chatting long distance

UNSW engineers have made a significant advance in quantum computing: they created ‘quantum entangled states’—where two separate particles become so deeply linked they no longer behave independently—using the spins of two atomic nuclei. Such states of entanglement are the key resource that gives quantum computers their edge over conventional ones.

The research is published in the journal Science, and is an important step toward building large-scale quantum computers—one of the most exciting scientific and technological challenges of the 21st century.

Lead author Dr. Holly Stemp says the achievement unlocks the potential to build the future microchips needed for quantum computing using existing technology and manufacturing processes.

The AI model that teaches itself to think through problems, no humans required

Artificial intelligence is getting smarter every day, but it still has its limits. One of the biggest challenges has been teaching advanced AI models to reason, which means solving problems step by step. But in a new paper published in the journal Nature, the team from DeepSeek AI, a Chinese artificial intelligence company, reports that they were able to teach their R1 model to reason on its own without human input.

When many of us try to solve a problem, we typically don’t get the answer straight away. We follow a methodical process that may involve gathering information and taking notes until we get to a solution. Traditionally, training AI models to reason has involved copying our approach. However, it is a long, drawn-out process where people show an AI model countless examples of how to work through a problem. It also means that AI is only as good as the examples it is given and can pick up on human biases.

Instead of showing the R1 model every step, researchers at DeepSeek AI used a technique called reinforcement learning. This trial-and-error approach, using rewards for , encouraged the model to reason for itself.

/* */