The top Tech Xplore articles of 2024

It was an interesting year for technology research as a team of human behaviorists at Aalto University, working with psychologist colleagues at the University of Helsinki Department of Psychology, found earlier this fall that a person’s intelligence limits their computer proficiency more than previously thought—they came to this conclusion by testing volunteers on their perception, thinking and remembering skills and also their ability to do common computer related tasks.

And this past winter, a team of materials scientists, physicists and engineers at University College London announced that they had achieved “near perfect” control of single atoms in a major advance toward building a truly useful quantum computer. They developed a new fabrication process involving the use of arsenic to reduce failure rates when building their computer and in so doing reduced such errors to near zero.

Also, a combined team of AI researchers and computer engineers from Shandong University and Nanjing University demonstrated a lightweight neural network that enabled realistic rendering of woven fabrics in real-time—doing so has been a long-time goal for computer graphics researchers. The researchers noted that their system works by encoding the patterns and parameters of fabrics as small latent vectors, which can later be interpreted by a decoder to produce realistic representations of various fabrics.

And last January, a team of engineers at the Columbia University School of Engineering and Applied Science made headlines when they announced that they had found, using an AI application, that not every fingerprint is unique. They made the discovery after training an AI system by feeding it 60,000 fingerprints and then asking it whether new pairs of prints belonged to the same person or to different people. Using such testing, the researchers found occasions of two people having nearly identical fingerprints.

Also, last January, an international team of physicists, chemists, and engineers, led by a group at the Georgia Institute of Technology, announced that they had created the first functional semiconductor made from graphene. The breakthrough, they reported, came when they figured out how to grow graphene on silicon carbide wafers using special furnaces. After perfecting the technique and adding other features, they found it could perform as a viable semiconductor.

And this past March, a team of materials scientists, engineers and chemists at Drexel University demonstrated their development of self-heating concrete. Two concrete slabs they laid down next to a parking lot have kept themselves snow-free for more than three years. The secret ingredient, they revealed, is paraffin, which releases heat when it undergoes a phase-change, such as from liquid to solid when the temperature dips below freezing.

Also, this past spring, a team of environmental scientists at the University of Pittsburgh revealed that they had found that if a way could be found to efficiently extract it, enough lithium could be pulled from normal wastewater to supply up to half of the demand—which is mainly batteries. They have found thus far that they can do it with approximately 90% efficiency. If a way could be found to extract lithium from wastewater nationally, they note, it would reduce the country’s reliance on foreign sources.

More recently, an international team of engineers led by a group at King Abdullah University of Science and Technology designed, built and demonstrated a new type of cooling system that works using gravity rather than electricity. The passive cooling system reflects thermal energy upward while simultaneously collecting cooled water pulled from the air at its bottom. The team notes that the device was built using inexpensive, easy-to-find parts.

Last spring, a team of German-American computer engineers led by a group at the University of Konstanz mapped the ownership of public network infrastructures in both democratic and authoritarian states worldwide to publicize how autocrats control the internet via state-owned service providers. The result, they claimed, was the production of global ownership of internet service providers; in essence, a kind of “atlas of internet surveillance” that anyone can use.

And this past fall, a group of researchers at the University of Toronto set out to explore how the use of LLMs during creative tasks affects the creativity of human users. In experiments they ran with volunteers, they discovered that use of such models can impair the ability of humans to think creatively, resulting in less varied and innovative ideas.

And this past summer, a team of roboticists at the University of Tokyo took a new approach to autonomous driving. After wondering if it might be more efficient to just put a robot in the driver’s seat of a car rather than automating the whole car, they built a robot and allowed it to do just that. They found that their robot could drive a car under easy circumstances but failed miserably when surprised—but they were also encouraged, noting that more work might pay off.

More recently, a team of electrical engineers at the University of Science and Technology of China iGAN Laboratory designed, built and demonstrated a promising three-terminal diode for wireless communication and optically driven computing—one that could both emit and detect light. They suggested it could open new possibilities for the development of highly performing wireless communication and light-driven computing systems.

Also, last April, leading computer scientists from around the world were asked to share their vision for the future of artificial intelligence—many of them thought it would resemble something like “The Borg” character of Star Trek fame. Most suggested the science would go that way because it is already on such a path. They noted that single AI systems that are really good at one thing are being connected to other AI systems that are really good at other things, giving them together a formidable presence.

And this past October, a team of engineers at AI inference technology company BitEnergy AI claimed that they had developed a method for reducing the energy needs of AI applications by approximately 95%. The proposed technique was based on using integer addition instead of complex floating-point multiplication when handling extremely large or small numbers, freeing up the machine to better handle numbers in between.

Also, last winter, a pair of researchers, writing for The Conversation, suggested that because very large solar farms can impact local weather, such technology is not limitless. Zhengyao Lu and Jingchao Long, contended that at some point, gigantic solar farms would start impacting weather to such an extent that at some point, the resulting clouds would block out so much sunlight that the solar farms could no longer operate—making them moot.

Also, last March, a team of researchers at the National Institute for Materials Science, in Japan, announced that they had designed, built and demonstrated the world’s first n-channel diamond metal-oxide-semiconductor field-effect transistor. They suggested the device represented a key step toward CMOS integrated circuits for harsh environment applications, as well as the development of diamond power electronics.

And this past fall, a team of engineers at the University of Massachusetts Amherst announced that they had invented a new way to align 3D semiconductor chips by shining a laser through concentric metalenses patterned on the chips to produce a hologram. They suggested their invention could lower the cost of producing 2D semiconductor chips, enable 3D photonic and electronic chips, and perhaps pave the way for other low-cost, compact sensors.

Also, just last month, a team of engineers at MIT announced that they had developed a new algorithm capable of revealing causal links in complex systems. Such a system, they note, could prevent claims of causal links in research efforts in almost any field. It could also perhaps help settle political, economic, or even personal arguments or debates about any topic where the reason something happened is not always obvious.

And last May, a team of engineers at The Australian National University proposed a new way to desalinate water that does not require the use of electricity. The new approach would instead use low-grade heat from sunlight, or heat that is a byproduct of an industrial process. It would then utilize thermodiffusion (a phenomenon where salt transfers to the colder side of a smooth temperature gradient) to remove the salt. The team also noted that the water would remain in a liquid phase throughout the process.

Also, this past month, a team of electrical engineers, economists and computer scientists from MIT, Harvard, Tisch University and Cornell University insisted that despite its impressive output, generative AI does not have a coherent understanding of the world. They used AI-based mapping applications as examples. They found that despite such apps being able to direct a driver to a location, they were not able to demonstrate any knowledge about the city where the driving was taking place, nor show that they even understood what driving meant.

And last spring, a team of chemists and engineers at The Korea Advanced Institute of Science and Technology announced that they had developed a sodium battery that was capable of rapid charging in just a few seconds. They described how their innovative hybrid energy storage system integrated anode materials typically used in batteries with cathodes suitable for supercapacitors. Such a combination, they noted, allowed the device to achieve both high storage capacities and rapid charge–discharge rates, positioning it as a viable next-generation alternative to lithium-ion batteries.

Also, over the summer, a team of computer scientists at ETH Zurich announced that they had developed the fastest possible flow algorithm. They explained that the algorithm involved use of what is known as a network flow algorithm, which tackles the question of how to achieve the maximum flow in a network while simultaneously minimizing transport costs.

And in October, an international team of engineers, meteorologists and environmentalists reported that they had found evidence showing that rooftop solar panels impact temperatures during both the day and night in cities. They created simulations showing how heat from widespread use of solar panels can increase temperatures in a given area. They noted that on a large enough scale, such heat could have an impact on the local weather.

And this past June, a team of engineers at Princeton University working with colleagues at UCLA showed how common plastics could passively cool and heat buildings with the seasons—such plastics, they showed, could restrict radiant heat flows between buildings and their environment to specific wavelengths. They also noted that coatings engineered from other common materials could achieve energy savings and thermal comfort that goes beyond what traditional building envelopes can achieve.

And finally, last March, a team of engineers at the University of Waterloo announced that they had created a new technology that could remove harmful nanoplastics from contaminated water with 94% efficiency. They explained how they used epoxy, a waste polymer that cannot be reused or reprocessed and often ends up in landfills, to build their system. Using a process called thermal decomposition, they converted epoxy into activated carbon, which could then be used to remove nanoplastics.

© 2024 Science X Network