Tech

Nearly every person tested in a massive U.S. study had PFAS, known as “forever chemicals,” circulating in their blood. Researchers examined more than 10,500 blood samples and found at least one PFAS compound in 98.8% of them.⁠ ⁠ The chemicals were rarely found alone. About 98.5% of people carried multiple PFAS at the same time, often in recurring combinations that scientists say could behave differently inside the body than single chemicals on their own.⁠ ⁠ PFAS are synthetic compounds designed to resist water, heat, grease, and stains. They have been used for decades in nonstick cookware, waterproof clothing, food packaging, electronics, carpets, firefighting foam, and industrial manufacturing.⁠ ⁠ Their biggest advantage is also what makes them dangerous. PFAS barely break down in nature, allowing them to build up in rivers, soil, wildlife, drinking water, and eventually human blood.⁠ ⁠ One chemical detected in nearly every sample, PFHxS, has already been associated in earlier studies with liver, immune system, and thyroid problems. Other PFAS, including PFOA and PFOS, have been linked to high cholesterol, fertility issues, weakened immunity, and certain cancers.⁠ ⁠ Scientists say the study highlights a growing problem in environmental health research. Most safety testing still looks at PFAS one chemical at a time, even though real-world exposure almost always involves mixtures of several compounds interacting together.⁠ ⁠ Researchers still do not know the full long-term effects of carrying these chemical combinations. But the findings make one thing clear: exposure to forever chemicals is no longer unusual. It has become part of everyday human existence.⁠ ⁠ Source: 10.1080/15459624.2025.2601605

Scientists in Hong Kong and China have created a new kind of battery using a water-based liquid similar to the mineral brine used to make tofu, and it may solve some of the biggest problems with modern energy storage. In laboratory tests, the battery survived more than 120,000 charge cycles, far beyond the lifespan of most lithium-ion batteries used today.⁠ ⁠ The key difference is the liquid inside the battery. Many rechargeable batteries rely on harsh chemicals that slowly damage their own parts over time and can become dangerous if they leak, overheat, or are thrown away improperly. This new design uses magnesium and calcium salts mixed into water at a neutral pH, making the battery nonflammable, less toxic, and much gentler on its internal components.⁠ ⁠ The team also built a new negative electrode from a carbon-rich material filled with tiny pores that help charged particles move in and out smoothly during charging. That structure allowed the battery to keep working with surprisingly little wear, even after years’ worth of simulated use. The researchers say the chemistry stayed stable because the materials avoided the corrosive reactions that normally shorten a battery’s life.⁠ ⁠ The battery still cannot store as much energy as the lithium-ion packs used in phones or electric cars, so it is not ready to replace them. But for giant backup systems connected to solar farms, wind turbines, factories, or electrical grids, the technology could be extremely useful. A safer battery that lasts dramatically longer and creates less environmental waste could become an important part of the world’s shift toward cleaner energy.⁠ ⁠ Source: 10.1038/s41467-026-69384-2

Sam Altman’s World wants to solve the internet’s deepfake problem by turning your iris into a fingerprint of your humanity.⁠ ⁠ The company, formerly known as Worldcoin, uses Orb devices to scan the colored ring around the eye, then converts that unique pattern into a World ID. The pitch is simple: prove you are a real person online without attaching the credential to your name or address, creating a portable layer of identity for an AI-saturated internet.⁠ ⁠ Now that system is moving into Tinder and Zoom. Tinder users can get a verified-human badge, while Zoom’s “Deep Face” tool checks a person’s Orb verification image, a live selfie, and the video frame other participants see on screen. The goal is to stop bots, romance scammers, and AI impersonators before they become nearly indistinguishable from real people, reshaping how trust is signaled across digital platforms.⁠ ⁠ The need is real. Dating apps are flooded with fake profiles, and deepfake fraud has already caused major losses, including a 2024 Hong Kong case where a worker sent $25 million after joining a video call filled with AI-generated colleagues posing as trusted executives.⁠ ⁠ But World’s answer comes with heavy baggage. Its early crypto-linked rollout drew criticism over recruitment tactics, especially in poorer countries. Regulators in Kenya and Europe have pushed back, and a Bavarian privacy ruling ordered the deletion of certain iris codes over data protection concerns.⁠ ⁠ World says nearly 18 million people have already verified across 160 countries. The deeper question is whether the future internet will require proof of humanity, and whether that proof should come from a private company scanning the body’s most permanent (and private) identifiers.⁠

The race to replace human workers with AI is accelerating fast, and one Chinese court has just stepped in to try and bring the impending doom to a halt.⁠ ⁠ The case centered on Zhou, a quality assurance worker at a tech firm in Hangzhou who reviewed outputs from large language models, the AI systems behind tools like ChatGPT. After the company automated parts of his role, it tried to move him into a lower position with a 40 percent pay cut. When he refused, it fired him.⁠ ⁠ The Hangzhou Intermediate People’s Court ruled the dismissal unlawful. Judges said AI adoption did not qualify as a “major change in objective circumstances,” the legal standard that can sometimes allow contracts to be ended. In plain terms, choosing to automate is a business strategy, not an uncontrollable crisis or unavoidable disruption.⁠ ⁠ The ruling does not stop companies from using AI. Instead, it says they must negotiate fairly, offer reasonable reassignments, protect worker rights, and prioritize retraining when technology reshapes jobs. A similar Beijing case involving an automated map-data role reached the same conclusion.⁠ ⁠ As companies worldwide race to replace human tasks with generative AI, China’s courts are signaling that automation cannot simply become a legal shortcut for discarding people. It suggests a future where innovation is forced to coexist with accountability, rather than override it entirely.⁠

Biotech company Colossal Biosciences is pushing its most ambitious idea yet: bringing back the bluebuck, an antelope hunted to extinction over 200 years ago. Once native to South Africa, the species vanished rapidly after colonial expansion, leaving behind only a handful of museum specimens and scattered records.⁠ ⁠ Now, those fragments are enough to rebuild its genetic blueprint. Scientists have sequenced bluebuck DNA and compared it to its closest living relative, the roan antelope. By editing the roan’s genome, they aim to recreate key bluebuck traits, then develop an embryo that could be carried by a surrogate mother. The process blends cloning, gene editing, and IVF into a single pipeline designed to reverse extinction at the molecular level.⁠ ⁠ The technical leap is massive. Researchers are narrowing millions of genetic differences down to the few that actually shape the animal’s appearance and biology. At the same time, they’ve developed new tools like ovum pick-up, a precision method for collecting egg cells, and lab-grown stem cells that can be turned into different tissues without testing on live animals.⁠ ⁠ Colossal says the goal isn’t just revival, but impact. Nearly a third of the world’s antelope species are threatened, and the same genetic tools could help boost biodiversity, strengthen fragile populations, and restore ecosystems where species have disappeared.⁠ ⁠ Still, the concept remains controversial. Critics argue these animals may be engineered approximations, not true returns, and question whether ecosystems can support them. The technology is advancing quickly, but whether it reshapes conservation or complicates it is still unfolding.⁠

NASA wants to start a fire on the Moon because future lunar homes may depend on knowing if it’s possible.⁠ ⁠ The mission is called Flammability of Materials on the Moon, or FM2. Planned for late 2026, it would send a sealed robotic chamber to the lunar surface aboard an uncrewed Commercial Lunar Payload Services lander, then ignite four solid fuel samples under tightly controlled conditions.⁠ ⁠ Fire changes when gravity changes. On Earth, hot gases rise, pulling fresh oxygen toward a flame and helping create its familiar teardrop shape. In microgravity, flames can become slow, round blobs that feed on nearby airflow. The Moon sits between those extremes, with about one-sixth of Earth’s gravity.⁠ ⁠ That partial gravity may be dangerous in unexpected ways. NASA researchers warn that materials considered only marginally nonflammable on Earth could burn more easily on the Moon, because weaker airflow may keep feeding oxygen without blowing the flame out. Future oxygen-rich lunar habitats could make that risk even sharper.⁠ ⁠ FM2’s cameras, radiometers, and oxygen sensors would track flame shape, spread rate, heat output, gas changes, and oxygen use over longer periods than drop towers or aircraft tests can provide. The data could reshape how NASA chooses fabrics, wiring, wall panels, spacesuits, insulation, and habitat materials for Artemis missions.⁠ ⁠ Before humans build lasting outposts beyond Earth, engineers have to relearn one of civilization’s oldest technologies: fire, inside a world where even a familiar flame may behave like something entirely new.⁠

Atari’s most infamous failure didn’t just flop, it was hauled out of a Texas warehouse, trucked into the New Mexico desert, crushed under concrete, and left there long enough to become folklore.⁠ ⁠ In 1983, after Atari’s business collapsed under the weight of oversupply, weak sales, and a market that had become overcrowded, the company dumped roughly 700,000 unsold cartridges and hardware into a landfill near Alamogordo. E.T. the Extra-Terrestrial became the symbol of that collapse, partly because the game was rushed into existence in about five weeks to hit the 1982 holiday season. The result was confusing, poorly received, and quickly returned by players, turning a blockbuster movie tie-in into one of the most notorious product failures of the era.⁠ ⁠ But the burial was never just about one bad game. Reports from the time, later interviews, and the 2014 excavation all showed a broader mix of titles, including Pac-Man, Centipede, Raiders of the Lost Ark, and other Atari material. What made the story endure was not only the scale of the dump, but the secrecy around it. Conflicting accounts, concrete poured over the site, and decades of rumor turned an act of corporate disposal into an urban legend.⁠ ⁠ Then archaeology turned that legend into documented history. In 2014, a Microsoft-backed documentary crew and archaeologists excavated a small slice of the landfill and recovered about 1,300 cartridges from 59 different games. Some were catalogued, some auctioned, and one E.T. cartridge ended up in the Smithsonian.⁠ ⁠ The story still stands out because it captures a moment when one of tech’s biggest companies moved too fast, misread demand, and left behind physical proof of the crash. Decades later, those buried cartridges still offer a rare, concrete record of how quickly a booming industry can unravel.⁠

A teenage inventor in Virginia may have found a smarter way to fight one of the most stubborn pollutants in modern water. Mia Heller, 18, created a filtration system that removes about 95.5% of microplastics from drinking water, using a design meant to be more affordable and less maintenance-heavy than many conventional filters. At a time when concern over invisible contaminants is growing, her idea stands out for pairing scientific precision with real-world practicality.⁠ ⁠ Her system replaces disposable membranes with ferrofluid, a magnetic oil-based liquid. As contaminated water moves through the device, the ferrofluid binds to microplastic particles. A magnetic field then pulls that material out, separating the pollution from the water. In simple terms, the liquid acts like a targeted collector for plastic fragments too small to easily trap with ordinary methods, offering a more elegant alternative to filters that clog, wear out, and need frequent replacement.⁠ ⁠ What makes the design especially striking is that it works as a closed-loop system. Heller’s current prototype uses three modules, filters about one liter at a time, and recovers most of the ferrofluid for reuse, cutting waste while reducing the need for constant replacement parts. Her testing found the device recycled 87.15% of the ferrofluid while maintaining strong microplastic removal, a combination that points to both efficiency and sustainability.⁠ ⁠ The idea grew out of a local problem. After learning about water quality issues in her community and watching her family deal with a high-maintenance home filter, Heller set out to build something more practical for everyday use. She has said the long-term goal is a consumer-ready system that could fit under a kitchen sink.⁠ ⁠ The bigger story is that a solution to a global pollution problem may come from rethinking the mechanics of cleanup itself, turning magnetism, fluid dynamics, and low-waste engineering into a tool that could make safer water more accessible.⁠

Gas prices became collateral damage in a geopolitical gamble, and Donald Trump made clear he can live with that. As oil markets jolted from the expanding U.S.-Iran conflict, he told Reuters he was unconcerned about higher prices at the pump, arguing the military operation mattered more than a temporary rise in fuel costs.⁠ ⁠ That stance lands at a moment when energy markets are acting like a real-time stress sensor for global conflict. Since the war began, oil prices have jumped sharply as traders priced in the risk of disrupted Middle East supply, especially around the Strait of Hormuz, the narrow shipping corridor that carries a huge share of the world’s seaborne crude. Gasoline followed quickly, with AAA reporting a noticeable weekly increase in the U.S. national average.⁠ ⁠ Behind the public confidence, the administration appears far less relaxed. Reuters reported that White House officials, including Chief of Staff Susie Wiles and Energy Secretary Chris Wright, have been in contact with oil executives while internal teams consider ways to contain the shock. Options under discussion reportedly include a federal gas-tax holiday, looser summer fuel rules that would allow higher ethanol blends, and other market-calming steps.⁠ ⁠ Trump, however, ruled out one of the biggest emergency tools available, a release from the Strategic Petroleum Reserve, at least for now. The White House is instead betting that the conflict, and the fuel-price spike tied to it, will be brief enough to avoid lasting political damage, unlikely as that is.⁠

In China, researchers and startup Magicpen Bio have engineered plants to shine in the dark by inserting light-producing genes from fireflies and bioluminescent fungi into plant cells. The result is a soft, visible glow generated by the plants themselves, not by bulbs, batteries, or wires.⁠ ⁠ The team says it has already created more than 20 glowing species, including orchids, sunflowers, roses, lilies, and chrysanthemums. Founder Li Renhan has described years of trial and error, including hundreds of technical iterations to improve the enzymes that drive the light-producing reaction and loosen the plants’ own genetic limits. The plants were recently shown publicly at the Zhongguancun Forum, where visitors saw illuminated flowers glowing without any external power source.⁠ ⁠ These plants need water and fertilizer, not electricity, which gives them appeal for parks, tourism zones, and other public spaces where conventional lighting carries an energy and emissions cost. Researchers frame the concept as a low-carbon complement to urban lighting, especially in places where a gentle biological glow could replace harsher artificial light.⁠ ⁠ This work is part of a broader push to turn plants into functional technology. Other Chinese researchers recently built sun-charged glowing succulents using synthetic materials, while MIT has spent years developing nanobionic plants that use luciferase, the same enzyme that helps fireflies glow. What once looked like decorative biotech is starting to resemble a new design language for cities, where illumination grows instead of being switched on.⁠

The race toward Q-Day continues. For years, quantum computers were framed as a distant threat to modern encryption, the math that protects everything from bank logins to private messages. Now that timeline looks less comfortable, not because the machines are fully ready, but because the science behind breaking today’s locks is advancing faster than expected.⁠ ⁠ The core issue is simple. Quantum computers process information with qubits, which can handle certain problems far more efficiently than conventional machines. That matters because widely used systems like RSA and elliptic-curve cryptography rely on mathematical challenges that are extremely hard for ordinary computers, but potentially much easier for a powerful enough quantum one using Shor’s algorithm.⁠ ⁠ What changed is the scale. New research from Google suggests elliptic-curve encryption, which underpins many secure communications systems and major cryptocurrencies, could be cracked with fewer than half a million physical qubits, a sharp drop from earlier estimates. A separate preprint from researchers tied to Caltech, Berkeley, and Oratomic argues that neutral-atom designs could push the requirement even lower for some attacks, with certain Bitcoin-related targets potentially vulnerable at around 26,000 qubits.⁠ ⁠ That does not mean the internet collapses tomorrow. Current systems are still far from those thresholds, and some experts caution that timelines remain uncertain. But the warning is no longer theoretical. Hardware is improving, algorithms are getting leaner, and attackers do not need to break data today if they can steal it now and decrypt it later.⁠ ⁠ The real story is the migration already underway. Post-quantum cryptography exists, standards are being rolled out, and the future of digital trust may depend on how quickly the world replaces the locks before the lockpick arrives.⁠

Electronics fail fast when heat rises. Most chips begin to break down above about 200 degrees Celsius, which has long limited everything from spacecraft and geothermal tools to high-performance industrial sensors. Now, researchers at the University of Southern California say they have built a memory device that keeps working at 700 degrees Celsius, hotter than molten lava and far beyond the range of conventional electronics.⁠ ⁠ The device is a memristor, a tiny component that can both store data and help perform computations. Its structure is simple but unusual: tungsten on top, hafnium oxide in the middle, and graphene, a one-atom-thick layer of carbon, on the bottom. That material stack turned out to be the key.⁠ ⁠ At extreme temperatures, standard devices usually die because metal atoms drift through the insulating layer, reach the other side, and create a permanent short circuit. Graphene stops that. The researchers found that tungsten atoms cannot properly bond to the graphene surface, so they move away instead of forming the connection that destroys the device.⁠ ⁠ The result is a chip that held data for more than 50 hours at 700 degrees, survived over one billion switching cycles, ran at just 1.5 volts, and switched in tens of nanoseconds. The team says 700 degrees was not the failure point, it was simply as hot as their equipment could test.⁠ ⁠ That opens the door to electronics that could survive Venus-like conditions, deep-earth drilling, and nuclear systems. It also matters for AI, because memristors can handle matrix multiplication, the math behind most modern AI workloads, with far less energy than standard chips.⁠ ⁠ Source: 10.1126/science.aeb9934