AusScot Computers Services

https://m.facebook.com/story.php?story_fbid=1439235548235028&id=100064457901208

Researchers built the world's first complete working CMOS computer without a single atom of silicon inside it, using materials that are just one atom thick. 💻🔬
Penn State researchers built the world's first functional CMOS computer constructed entirely from atom-thin 2D materials, replacing silicon with molybdenum disulfide and tungsten diselenide in a working integrated circuit, published in June 2025 and highlighted as a landmark materials engineering milestone through 2026. Silicon has dominated computing since the 1950s and is now rapidly approaching its physical miniaturization limits. Atom-thin 2D materials can be made into transistors far smaller than silicon allows while consuming dramatically less power, because quantum effects at single-atom thickness give these materials electrical properties silicon cannot match at equivalent scales. The Penn State team demonstrated full CMOS logic operations — the fundamental building block of every computer — in a device that fits on a substrate smaller than a fingernail.
Silicon transistors are currently around 2 nanometres in size and approaching the atomic scale limit. A computer that runs on materials just one atom thick bypasses that physical wall entirely and opens a new trajectory for computing miniaturization that could continue for decades beyond where silicon must stop.
-
News Source: Penn State / ScienceDaily, "Atom-Thin Tech Replaces Silicon in the World's First 2D Computer" (June 2025, highlighted through 2026)
-

https://m.facebook.com/story.php?story_fbid=1436054121886504&id=100064457901208

New research showed that AI systems do not need massive amounts of training data when their architecture is redesigned to better resemble how biological brains actually process information. 🤖🧠
Researchers published findings on January 4, 2026, via ScienceDaily, demonstrating that AI models redesigned to more closely mirror the structure of biological neural networks showed significant behavioral improvements, including faster learning from fewer examples and better generalization to new situations, compared to conventionally structured models trained on orders of magnitude more data. Rather than relying on sheer volume of training data and raw computational scale, the brain-inspired architectures leverage the same principles of sparse connectivity, local processing, and temporal dynamics that make biological brains extraordinarily efficient. The work directly challenges the dominant assumption in AI development that more data and bigger models are the primary drivers of better AI performance.
If AI systems can learn more efficiently from less data by mimicking brain structure more closely, the energy and infrastructure demands of AI development could fall dramatically, making powerful AI accessible far beyond the handful of companies currently able to afford training the world's largest models.
-
News Source: ScienceDaily, "New research shows that AI doesn't need endless training data to start acting more like a human brain" (January 4, 2026)
-

https://m.facebook.com/story.php?story_fbid=1434176815407568&id=100064457901208

Scientists just transported the most expensive, volatile, and rarest substance on Earth in the back of a truck, and celebrated with Champagne when it arrived intact. ⚛️🚚
On March 24, 2026, physicists at CERN's Antimatter Factory near Geneva loaded 92 antiprotons into a specially built transportable trap, a one-metric-ton box kept at minus 269 degrees Celsius using supercooled magnets and a near-perfect internal vacuum, placed it on a truck, and drove it six miles around CERN's campus for 30 minutes. As professor Alan Barr of Oxford University explained: "The moment these antimatter particles come into contact with normal matter, they annihilate each other. They just vanish in a puff of light." The trap held. Every antiproton arrived undisturbed. The goal is eventually to drive antiprotons eight hours to Heinrich Heine University in Düsseldorf, where quieter lab conditions could allow measurements 100 to 1,000 times more precise than anything achievable inside CERN's own magnetic interference.
Why does this matter? Antimatter holds the answer to why the universe exists at all. The Big Bang should have created equal amounts of matter and antimatter that annihilated each other completely. Something stopped that from happening, and finding out what could be the single biggest discovery in the history of physics.
-
News Source: CNN / PBS NewsHour / Scientific American, "CERN takes antimatter out of the lab for the first time ever" (March 25, 2026)
-

A newly constructed data center in Georgia reportedly consumed approximately 29 million gallons of water before utility billing systems were fully activated, even as local residents were being urged to conserve water during periods of high demand. The incident has intensified concerns over the growing resource requirements of AI infrastructure.

Modern hyperscale data centers require enormous quantities of water to cool thousands of high-performance servers that operate continuously. Some large facilities consume millions of gallons each day, depending on cooling technology, weather conditions, and computing workloads.

The case has renewed calls for greater transparency regarding water use, utility agreements, and environmental reporting as AI infrastructure expands. Communities are increasingly asking technology companies to adopt recycled water systems, closed-loop cooling technologies, and more efficient liquid-cooling methods to reduce pressure on local water supplies.

As artificial intelligence continues driving unprecedented data center construction, water availability is becoming as important as electricity in determining where future facilities can be built.

Large data centers consume vast amounts of electricity, much of which is ultimately released as heat.

Researchers have found that clusters of data centers can contribute to localized warming in surrounding areas, particularly where many facilities are concentrated.

However, claims that data centers are heating regions up to 6 miles away and affecting 343 million people should be treated cautiously, as impacts vary widely depending on climate, cooling technology, facility size, and geography.

The growing energy demands of AI are increasing interest in more efficient cooling systems and heat-recovery technologies.

They didn't trap the foxes. They vaccinated them.

Switzerland eliminated rabies by hiding vaccine doses inside bait and scattering them in the wild. Foxes ate the bait, vaccinated themselves, and spread protection across the country.

It worked. Cases dropped to zero. A nation protected itself—by protecting its foxes. 🦊💊🇨🇭✅