Footage showing two players in a locally-hosted game is “really hacky,” but it works.
Get the latest international news and world events from around the world.
Engineers invent wireless transceiver that rivals fiber-optic speed
A new transceiver invented by electrical engineers at the University of California, Irvine boosts radio frequencies into 140-gigahertz territory, unlocking data speeds that rival those of physical fiber-optic cables and laying the groundwork for a transition to 6G and FutureG data transmission protocols.
To create the transceiver, researchers in UC Irvine’s Samueli School of Engineering devised a unique architecture that blends digital and analog processing. The result is a silicon chip system, comprising both a transmitter and a receiver, that’s capable of processing digital signals significantly faster and with much greater energy efficiency than previously available technologies.
The team from UC Irvine’s Nanoscale Communication Integrated Circuits Labs outline its work in two papers published this month in the IEEE Journal of Solid-State Circuits. In one, the engineers discuss the technology they call a “bits-to-antenna” transmitter, and in the second, they cover their “antenna-to-bits” receiver.
AI models mirror human ‘us vs. them’ social biases, study shows
Large language models (LLMs), the computational models underpinning the functioning of ChatGPT, Gemini and other widely used artificial intelligence (AI) platforms, can rapidly source information and generate texts tailored for specific purposes. As these models are trained on large amounts of texts written by humans, they could exhibit some human-like biases, which are inclinations to prefer specific stimuli, ideas or groups that deviate from objectivity.
One of these biases, known as the “us vs. them” bias, is the tendency of people to prefer groups they belong to, viewing other groups less favorably. This effect is well-documented in humans, but it has so far remained largely unexplored in LLMs.
Researchers at University of Vermont’s Computational Story Lab and Computational Ethics Lab recently carried out a study investigating the possibility that LLMs “absorb” the “us vs. them” bias from the texts that they are trained on, exhibiting a similar tendency to prefer some groups over others. Their paper, posted to the arXiv preprint server, suggests that many widely used models tend to express a preference for groups that are referred to favorably in training texts, including GPT-4.1, DeepSeek-3.1, Gemma-2.0, Grok-3.0 and LLaMA-3.1.
A new atlas could help guide researchers studying neurological disease
Functioning brain cells need a functioning system for picking up the trash and sorting the recycling. But when the cellular sanitation machines responsible for those tasks, called lysosomes, break down or get overwhelmed, it can increase the risk of Alzheimer’s, Parkinson’s, and other neurological disorders.
“Lysosomal function is essential for brain health, and mutations in lysosomal genes are risk factors for neurodegenerative diseases,” said Monther Abu-Remaileh, a Wu Tsai Neuro affiliate and an assistant professor of chemical engineering in the Stanford School of Engineering and an assistant professor of genetics in the Stanford School of Medicine.
The trouble is, scientists aren’t sure exactly how lysosomes do their work, what’s going wrong with lysosomes that leads to neurodegeneration—or even in which cell types neurodegenerative disease begins. There might even be other lysosomal disorders yet to be discovered.
New code connects microscopic insights to the macroscopic world
In inertial confinement fusion, a capsule of fuel begins at temperatures near zero and pressures close to vacuum. When lasers compress that fuel to trigger fusion, the material heats up to millions of degrees and reaches pressures similar to the core of the sun. That process happens within a miniscule amount of space and time.
To understand this process, scientists need to know about the large-scale conditions, like temperature and pressure, throughout the target chamber. But they also want detailed information about the material—and the atoms—contained within. Until now, computer models have struggled to bridge that gap across the wide range of conditions encountered in such experiments.
ATLAS confirms collective nature of quark soup’s radial expansion
Scientists analyzing data from heavy ion collisions at the Large Hadron Collider (LHC)—the world’s most powerful particle collider, located at CERN, the European Organization for Nuclear Research—have new evidence that a pattern of “flow” observed in particles streaming from these collisions reflects those particles’ collective behavior. The measurements reveal how the distribution of particles is driven by pressure gradients generated by the extreme conditions in these collisions, which mimic what the universe was like just after the Big Bang.
The research is described in a paper published in Physical Review Letters by the ATLAS Collaboration at the LHC. Scientists from the U.S. Department of Energy’s (DOE) Brookhaven National Laboratory and Stony Brook University played leading roles in the analysis.
The international team used data from the LHC’s ATLAS experiment to analyze how particles flow outward in radial directions when two beams of lead ions—lead atoms stripped of their electrons—collide after circulating around the 17-mile circumference of the LHC at close to the speed of light. The findings offer new insight into the nature of the hot, dense matter generated in these collisions—with temperatures more than 250,000 times hotter than the sun’s core. These extreme conditions essentially melt the protons and neutrons that make up the colliding ions, setting free their innermost building blocks, quarks and gluons, to create a quark-gluon plasma (QGP).
Growth chambers could enable reproducible plant-microbe data across continents
Harnessing the power of artificial intelligence to study plant microbiomes—communities of microbes living in and around plants—could help improve soil health, boost crop yields, and restore degraded lands. But there’s a catch: AI needs massive amounts of reliable data to learn from, and that kind of consistent information about plant-microbe interactions has been hard to come by.
In a new paper in PLOS Biology, researchers in the Biosciences Area at Lawrence Berkeley National Laboratory (Berkeley Lab) led an international consortium of scientists to study whether small plastic growth chambers called EcoFABs could help solve this problem.
Building on their previous work with microbe-free plants, the scientists used the Berkeley Lab-developed devices to run identical plant–microbe experiments across labs on three continents and got matching results. The breakthrough shows that EcoFABs can remove one of the biggest barriers in microbiome research: the difficulty of reproducing experiments in different places.
Study reveals why light-driven chemical reactions often lose energy before bond-breaking
Florida State University researchers have discovered a pathway within a certain type of molecule that limits chemical reactions by redirecting light energy. The study could enable development of more efficient reactions for pharmaceuticals and other products.
The researchers examined ligand-to-metal photocatalysts. Ligands are a molecule bound to a larger molecule; in this case, to a metal. Photocatalysts are materials that use light to accelerate a chemical reaction. Theoretically, these molecules should be readily able to harness light energy toward chemical reactivity. But in experiments, chemists only found inefficient reactions.
The FSU research, published in the Journal of the American Chemical Society, shows why: The molecule quickly moves into a less energetic state before the absorbed energy can break chemical bonds. The energy is drained too quickly into the wrong place, so bond-breaking is limited.
Molecular surgery: ‘Deleting’ a single atom from a molecule
Inserting, removing or swapping individual atoms from the core of a molecule is a long-standing challenge in chemistry. This process, called skeletal editing, can dramatically speed up drug discovery or be applied for upcycling of plastics. Consequently, the field is witnessing a surge of interest spanning from fundamental chemical research to applications in the pharmaceutical industry.
A group of researchers have now extended the scope of skeletal editing to the scale of just a single molecule. Such a level of precision in skeletal editing is unprecedented, and this may open a new route to obtain elusive molecules.
The team of researchers are active at Chalmers University of Technology, Sweden; IBM Research Europe—Zurich, Switzerland; and CiQUS at the University of Santiago de Compostela, Spain. In a recent article published in the Journal of the American Chemical Society, they demonstrate how, in a controlled manner, they can selectively remove a single oxygen atom from an organic molecule using the sharp tip of a scanning probe microscope.
Single enzyme streamlines production of all four RNA building blocks
A single enzyme that can generate all four nucleoside triphosphates, the building blocks of ribonucleic acid (RNA), has been identified by researchers at the Institute of Science Tokyo. The study was published online in the journal Nature Communications.
By using polyphosphate as a phosphate donor, the enzyme efficiently converts inexpensive nucleotide precursors into the active forms required for RNA synthesis. Overall, the method dramatically simplifies the process of nucleotide production—offering a low-cost, efficient option for the in vitro synthesis of RNA.