You are here

The biggest scientific discoveries of 2023

Every year, scientists make breakthroughs in all manner of important fields, and 2023 was no different. In a society in which everyone has powerful computers in their pockets, some cars can drive themselves (yet still not fly), and merely mentioning a product to the right device can cause it to show up on your doorstep, it's getting harder and harder to be surprised — but the scientific strides made in 2023 may still have some jaws on the floor and eyes popping comically out of heads.

With the advent of the artificial intelligence-powered chatbot ChatGPT, AI was big news in '23, and that particular field generated even more fascinating innovations that you may not yet be aware of. But pretty much all of the scientific disciplines took their turns kicking butt and taking names throughout the year, from paleontology to biology to computing, and many of the advances that were made could have profound implications for the future. So, without any further ado, let's dive into them — the biggest scientific discoveries of 2023. 


Archaeologists have been discovering cave drawings since time immemorial, but identifying anything resembling an actual system of writing used by our ancient ancestors has remained elusive. Some indications have been found, however; for well over a century, scientists have documented what appears to be a set of markings, accompanied by drawings of animals as old as 37,000 years, appearing in cave systems across Europe. Until recently, what these markings might mean was anyone's guess — but in January 2023, a study published in the Cambridge Archaeological Journal may have finally pinned down their purpose.

Researchers compared the symbols to the birth cycles of the animals depicted in the drawings and the 13 annual lunar cycles, and astonishingly, a pattern was found. It indicated that the markings might actually be an extremely early — perhaps the earliest — notational system, one which pertains to the mating and birth cycles of species that would have been our ancestors' prey. In particular, one symbol resembling the letter "Y" seemed to correlate to birth, the first known instance of such a correlation in cave markings. 

It's not simply an interesting discovery; it suggests that early humans may have been capable of more advanced cognition than previously thought. "This is a really big deal cognitively," said Ben Bacon, a researcher and co-author of the study. "We're dealing with a system that has intense organization, intense logic to it." He further explained that the system's basis on lunar cycles made it "universal ... if you migrate from the Pyrenees to Belgium, you can just use the same calendar."


Back in 2012, Japanese biologist Shinya Yamanaka won a share of the Nobel Prize in Physiology or Medicine. It was for his part in discovering a series of proteins capable of reconfiguring your average, everyday cells, turning them into super-versatile stem cells, which almost certainly have far more therapeutic uses than have been discovered yet. 

Over a decade later, a pair of teams from the United States reported on processes that employed these protein cocktails, known as "Yamanaka factors," in gene therapy treatments, that proved to have a very interesting effect on mice in the lab. "These studies," explained University of Washington scientist Matt Kaeberlein to the journal Science, "use reprogramming factors to reverse epigenetic changes that happen during aging" — an effect that has the potential to one day dramatically extend the lifespan of humans, or even turn back the clock on old age.

To be sure, the technology is still in its infancy (no pun intended), but the results of the twin studies are fascinating; one study (via Biorxiv), from the biotech company Rejuvenate Bio in San Diego, California, claimed to effectively double the life expectancy of elderly mice using the therapy. The other (via Cell), spearheaded by the somewhat controversial Harvard geneticist David Sinclair, lent credence to Sinclair's hypothesis that the effects of aging are due to a specific type of deterioration of DNA, reversing some of those effects by repairing the deterioration using Yamanaka factors. The tech has a long way to go before it can be applied to a potential fountain of youth — but these early results show that the potential is real.


If the phrase "liquid metal robot" may call to mind an image of actor Robert Patrick in a police uniform, but the versions presented by researchers at Pennsylvania's Carnegie Mellon University in January 2023 looked nothing like time-traveling shapeshifting cops; they were extremely tiny, activated to change states by magnetic fields, and capable of performing delicate tasks in both mechanical and medical capacities.

Composed of the liquid metal gallium and extremely tiny particles of an alloy, the robots are capable of supporting 30 times their weight when solid — but when exposed to magnetism, they turn to puddles that can be pushed, pulled, and stretched to perform tasks via manipulation of the magnetic field. According to Cell, these tasks included maneuvering a small light bulb to a circuit board, then attaching it by melting over it and providing a current; navigating to a foreign substance inside an artificial stomach, melting over it, and hauling it out; and transforming into the shape of a tiny man, then liquefying to slip between the bars of a tiny jail cell. (No, really. They actually made it do this.) 

A foolproof method of precisely tracking the little guys' location will have to be developed before they can be deployed in medical procedures — but when it comes to other uses, such as performing minor mechanical procedures on spacecraft, the tech is essentially ready to go.


When it comes to wireless networks and, well, electronics in general, a large part of developing greater speed and efficiency comes down to miniaturization. Simply put, the smaller electronic components can be made while retaining their power and efficiency, the better — and in the case of wireless networks, smaller signal wavelengths translate to faster data transmission. Given how tiny yet powerful electronic devices have become, though, it may not shock you to learn that miniaturization has pretty much hit a wall — which is where the concept of the "metadevice" comes into play, a concept which will potentially form the foundation of the sixth generation (6G) of wireless protocols.

Without going full egghead, the general principle is this: metadevices are basically conventional semiconductors adorned with "metastructures," which are contacts etched into their surfaces in microscopic patterns shorter than the device's wavelength output. These metastructures allow for manipulation of the electrical fields within the semiconductor, forcing it to output frequencies in the terahertz range — which, for those keeping score at home, is way, way faster than the gigahertz-range frequencies employed by common wireless devices.

As reported by EPFL, Ph.D. student Mohammad Samizadeh Nikooby — the co-author of the study published in Nature – said, "This new technology could change the future of ultra-high-speed communications, as it is compatible with existing processes in semiconductor manufacturing." The next step: the engineering of other electrical components compatible with this new wave of semiconductors.


3-D printing is a marvelous recent technology that has led to advances in architecture, irrigation, health care, and a slew of other fields. In a nutshell, 3-D printers use a layering process to "print" entire objects out of special "ink." But since its inception, it has suffered from a specific limitation: the ink. Due to the fact that it must behave as both a liquid and a solid, the end products of the 3-D printing process can often be too brittle or too malleable — but in February 2023, researchers at EPFL's School of Engineering published a study in Materials Today which details a novel solution to the problem.

That solution is living bacteria, Sporosarcina pasteurii, the critical ingredient in what the team calls "BactoInk." The special ink can be used in any standard 3-D printer, and no special printing process is needed — but the finished product, once treated with a solution, mineralizes and hardens over several days. Once the process is complete, the mineral content of the object stands at over 90 percent — rendering it far more durable than conventional 3-D printed objects. The team foresees potential uses in the medical field, in the restoration of delicate artworks such as sculptures and vases, and — perhaps most intriguingly — the possibility of 3-D printing artificial coral reefs, which could help to repair damaged marine ecosystems. 


The use of electrodes in medical technology, which are used to artificially stimulate nerves, muscles, and even the brain, can be akin to a double-edged sword — literally, in a way. No matter the materials they are made from, none are soft enough to completely mitigate the risk of damage to soft tissues, or the irritation and inflammation of those tissues, which can affect the performance of the electrode. 

In February 2023, though, researchers at Sweden's Linköping University published a study in Science that suggests that those issues may soon be a thing of the past. It describes the composition of a special gel that, when injected into the target area, metabolizes the body's own sugars to transform it into an ultra-soft, flexible electrode that carries all of the zap, so to speak, with none of the risk.

The team used the technique on two different organisms for different purposes: the soft electrodes were successfully grown in the hearts, tail fins, and brains of Zebrafish to stimulate those organs and muscles, and the process was used in leeches to stimulate a nerve. Further study is needed to determine how the electrodes will perform over the long run, and if there are any long-term issues involved in their use — but those initial results showed a great deal of promise, and the tech is even undergoing further refinement. As the study's co-author, Xenofon Strakosas, explained to Science News, "We're conducting experiments right now where we're trying to bind these materials directly on individual cells."


Octopuses are known to be some of the most intelligent creatures on the planet, but not much is known about how their brains work, and for pretty simple reasons: even the most seasoned marine biologist is going to have an issue trying to attach anything to an octopus, given how extremely squishy they are, as well having dextrous tentacles ideal for removing any attached monitoring equipment.

In March 2023, University of Naples researchers published a study in Cell outlining a way around this: They modified data loggers of the type normally used on birds, and surgically implanted them into three octopuses, which were then allowed to roam freely in their tanks. The result was the most accurate recording yet of the animals' brainwave patterns, some of which are of the type that have literally never been seen before.

Researchers described the kind of brain waves that are associated with memory consolidation in humans, and those which control sleep cycles in some other animals. They also described exceptionally slow, strong brain waves that were singularly unique, which were picked up both by single electrodes and, in other instances, by multiple electrodes in different regions of the animal's body. The research represents an exciting new technique for learning about the very nature of intelligence: "The problems that [octopuses] face are the same problems, but the solutions that they find are sometimes similar and sometimes different," explained researcher Tamar Gutnick to Science News. "And all of these comparisons teach us something."


Type 1 diabetes, when the body fails to produce sufficient amounts of insulin, has a strong inherited genetic component, and can be triggered by environmental factors not yet fully understood, per the American Diabetes Association. It causes the body's immune system to attack insulin-producing cells — but in March 2023, a team at ETH Zurich, a research university in Switzerland, came up with a brilliant method of heading the disease off at the pass.

Their study (via Advanced Materials) describes a tea bag-sized, implantable fuel cell that detects extra glucose as blood sugar levels rise. It then splits that glucose into its individual components, including protons, which it then utilizes to generate an electrical current. Combined with artificial cells within the implant, this current is employed to actually synthesize and secrete insulin. When the body's glucose level returns to normal, the process automatically terminates — allowing the body to self-regulate blood sugar levels in the same way that those of non-diabetics do. It's an ingenious bioelectric solution to an age-old health issue, and one that could be used to successfully treat people with diabetes the world over in the near future.


In March 2023, a study was published in Nature — after more than a year of review — that has stunning implications for the future of in vitro fertilization. Unless you're a biologist, you may not glean as much from its title: "Generation of functional oocytes from male mice in vitro." But, as explained by The Washington Post, the study represents a startling achievement. It describes a process that only achieved its goal one percent of the time, but what a goal it was to even see that level of success — the creation of healthy mouse pups from the cells of two male mice.

It must be understood that this has nothing to do with cloning, or anything of that sort — using a combination of techniques, skin cells from the tails of the mice were first converted into stem cells. Those highly versatile, malleable cells were then converted into female cells, which were used to produce egg cells. Those eggs were fertilized and implanted into female mice, and — in seven instances out of 630 attempts — grew into healthy mouse pups, which developed normally and were able to have pups of their own as adults. 

Now, that success rate needs some work, and it's still unclear if this technique could be ported to humans — but if it can, it could enable gay male couples to actually have kids of their own without the need for donor eggs, for example. The tech is also being eyed as a potential avenue to rescue endangered species from extinction using the cells of one male specimen.


In their quest to discover how life on Earth began, scientists have often looked to the stars — or, less poetically, to space rocks. In the past, meteorites have been found to contain probiotic compounds which, in conjunction with other compounds, can form the building blocks of life — but those meteorites passed through the Earth's atmosphere, leaving the possibility that they could have been contaminated. The question of whether rocks from the great void can contain such materials, then, has remained an open one — until March 2023, when analysis of samples collected by the Japanese probe Hayabusa 2 was completed, and a study based on the findings published in Nature.

Hayabusa 2 was sent to collect samples from near-Earth asteroid 162173 Ryugu, and said samples were indeed found to contain — in very small amounts, of course — uracil, a compound that is one of four critical components of RNA. Traces of niacin, also known as Vitamin B3, were also detected. Hokkaido University researcher Yasuhiro Oba confirmed that, theoretically, if such a body were to crash-land on a lifeless planet, these trace elements might eventually contribute to the formation of life — a conclusion that certainly lends some credence to the Panspermia hypothesis, which posits that life on Earth may have begun just this way.


You're probably aware that the prefix "nano-" means "really, really small." A nanowire network, then, is just what it sounds like — a network composed of tiny silver wires, so tiny as to be invisible to the naked eye individually, laid over each other like a mesh and interconnected with each other. In April 2023, a team of University of Sydney scientists published a study in Science describing an aspect of their nanowire network that was nothing short of amazing: it had demonstrated the ability to perform higher-level cognitive tasks, the likes of which were previously solely the domain of humans.

The network, in fact, acts in much the same way the brain does; the wires serve the same function as neurons, and points where they connect act as synapses. On a psychology test known as the "N-Back" test — in which the network was tasked with identifying images that it had previously viewed several steps back in a sequence — the network performed just as well as a human, and even "banked" the images in its memory, demonstrating the capacity for both short- and long-term memory processing. 

In a press release (via the University of Sydney), the study's co-author, Professor Zdenka Kuncic, explained that this form of "artificial intelligence" is a step beyond machine learning. "[We have] actually taken it one step further," she said, "and tried to demonstrate that nanowire networks exhibit some kind of cognitive function" — and apparently, they succeeded.


What if a computer could literally read and transcribe your thoughts? Okay, on the surface, that might sound absolutely terrifying. But, think about it for a second — what if, for example, Stephen Hawking had been able to wear a helmet that could simply have told us what he was thinking? Or, what if an author could compose entire passages, or entire novels, directly from their thoughts? If this sounds like the stuff of the far-flung future, then we must be pretty far-flung, because in May 2023, University of Austin researchers published a study in Nature describing an AI model that can do just that — and it's non-invasive, monitoring its subjects' brain waves via a simple MRI.

The study's co-lead, Alex Huth, described the tech as a huge leap forward over what had been accomplished previously — rather than just a few words or simple sentences, this AI can transcribe complex thoughts and imagery from brainwaves alone. Sure, it's not perfect — sometimes things get a bit lost in translation, with keywords or concepts altered or absent from the transcription, and the AI often parses out the meaning of what is being thought rather than the exact words. ("He came through the door" might translate to "He entered the room," for example.) 

However, for a non-invasive method, the results achieved by the team are astonishing — and when the method of brain imaging is scaled down from a clunky MRI machine, a portable version could completely change how we think about, well, thinking.


Launched on Christmas Day 2021, NASA's James Webb Space Telescope has returned so many mind-bending images, and made so many significant findings, that it's tough to pin down the most significant of them all. It has shown us a Milky Way-like galaxy millions of miles away, painted us the most brilliant picture of our universe we have ever seen, and given us new mysteries to solve concerning the earliest-formed galaxies in the universe. But of all of its observations, the one announced by NASA in July 2023 just might have the most profound implications.

The discovery came about during the Webb's examination of PDS 70, a planetary system some 370 light-years away. Its star is home to two Jupiter-like gas giants, located between two ginormous disks of gas and dust. Within the inner disk lies the region where Earth-like, rocky planets may be forming, already home to two "proto-planets" — and there the Webb found water vapor, perhaps the most important component in the potential development of life. 

Giulia Perotti, co-author of a paper on the finding, explained, "We've seen water in other disks, but not so close in and in a system where planets are currently assembling," adding, "We couldn't make this type of measurement before Webb" (via NASA). This may allow us greater insight into how Earth was formed, and how life came to populate it — and it may also offer us a sneak peek into the birth of a new, life-supporting solar system.


The development of a substitute for palm oil might not seem like that big of a deal, unless you are a staunch environmentalist. Here are the facts, though, that everyone should be aware of: palm oil is present in nearly half of all food and cosmetic products in existence. It is by far the world's most-produced vegetable oil — and its production is a major cause of deforestation, driving instability in ecosystems and threatening myriad species with endangerment or extinction. A viable substitute would be a literal breath of fresh air for the planet — and a team of scientists at Queen Margaret University in Edinburgh, Scotland, claim they have found one, calling it the "Holy Grail to replace palm oil."

Dubbed PALM-ALT, the substance is entirely plant-based and, its developers say, 70 percent more environmentally friendly than palm oil. In processed food, palm oil is odorless and tasteless, used to enhance texture and act as a preservative — and it is also loaded with saturated fat and calories, which PALM-ALT reduces by up to 80 percent, as reported by the BBC. "We've put it through some special sensory testing to see if a panel can tell the difference between our product and traditional palm shortening," explained Catriona Liddle, one of the team's lead developers, "and they can't." 

If the product is made widely available, this could spell excellent news for consumers — not to mention for the elephants, tigers, rhinos, and other species affected by the clear-cutting of African forests to grow the palm trees used in traditional palm oil production. 


It's one thing to extract DNA from fossilized or otherwise long-dead organisms, "Jurassic Park" style, as it can hold up well over long periods of time. RNA, though — a single-stranded copy of a section of DNA — is another story, as it was previously thought impossible to extract this more delicate molecule in the same fashion. In September 2023, though, a research team from Sweden's Centre for Palaeogenetics and SciLifeLab published a study in Genome Research, announcing that they had cracked the code for extracting RNA from ancient specimens — and in doing so, they opened up an avenue for gathering much more information about extinct species than has previously been available. 

The specimen in question came from a thylacine, a type of Tasmanian tiger that has been extinct for nearly a century — the last living specimen died in captivity in 1936. As the study's lead author, Emilio Mármol Sánchez, explained to CNN, "RNA gives you the chance to go through the cell, the tissues and find the real biology that has been preserved in time for that animal ... right before they died." 

This offers much more information about the animal's genetic makeup than was previously available — and while Mármol Sánchez emphasized that the goal of his team was not necessarily to bring about the means to revive the thylacine from extinction, access to more detailed genetic info could bolster such endeavors, which have been underway for some time. 


For the time being, computing still largely relies on traditional silicon processors, which have hit something of a ceiling in terms of speed, power, and energy efficiency. At least, they had — before IBM, the company that coined the term "PC" back in the computing dark ages, changed that with the invention of a next-generation chip that makes other processors look like, well, relics of those dark ages. It's called "Northpole," its architecture is inspired by that of the human brain, and it runs AI applications at unprecedented speeds thanks to an innovative design.

Simply put, the chip integrates the processor and memory into a single unit, which cuts down on the amount of data transfer between individual components. Cuts it down so much, in fact, that the resulting increase in energy efficiency and decrease in latency (or lag time, if you will) is enough to outperform even the most hardcore processors currently on the market. 

The team that developed Northpole put it up against the fastest units out there, including the latest iterations of the famously powerful Nvidia GPUs — and it smoked them all, to the tune of 22 times faster than the next fastest processors. The team cautioned that the chip is currently only suitable to run specialized AI processes — but that chaining multiple Northpoles together should be able to open it up to a broader range of applications, up to and including gigantic, language-based AI models like ChatGPT.


Next-generation computer processors are one thing, but quantum computers are different horses altogether. The technology is just beginning to be ready for prime time; IBM introduced its quantum computing model to the cloud in 2016, and released a unit in 2019 — the IBM Q System One — that is available to the consumer, albeit at an eye-watering price. 

In simplest terms, quantum devices use "qubits," the quantum equivalent of a digital bit, to process data. Qubits are bits made from a quantum mechanical system, like an electron — and rather than being locked into either one state or another, like a classical bit, a qubit can exist in both, or neither, or be 'entangled" with other qubits. In short, their processing power is extremely robust — and while most existing quantum computers are in the neighborhood of 20 to 50 qubits, a company called Atom Computing has designed one that is orders of magnitude greater.

In October 2023, the company announced it had achieved a system with a record-shattering 1,180 qubits — just a touch better than their previous system, which came online in 2021 and offered only 100. Today, quantum computing is not exactly feasible for practical applications due to its error rates — but as the number of qubits increases, so does the potential for error correction, and in announcing their achievement, Atom founder Ben Bloom explained to Ars Technica that he envisions "hundreds of thousands to millions of qubits in a single system" — a level of scalability which now actually seems within reach, and which could have profound implications for the future of cryptography, AI, medical research, and much more.


At an average surface temperature over 800 degrees Fahrenheit, Earth's neighbor Venus is hardly hospitable to anything resembling life. But scientists have long puzzled over whether this was always the case — and in October 2023, a study published in Nature by the Lunar and Planetary Institute in Houston, Texas, offered up new evidence that the planet may once have been much more Earth-like. The study suggests that the high levels of nitrogen and carbon dioxide in the atmosphere of Venus, which contribute greatly to its Hades-like surface temperature, could be explained by ancient plate tectonics — the continual reshaping of the planet's crust.

The study cited computer modeling suggesting the planet's current atmosphere could be explained by an ancient period of tectonic activity, followed by the crust seizing into its current, stable "stagnant lid" model. If true, then during that ancient period, the planet might have had far cooler temperatures, and even liquid water — and theoretically could have supported life. 

A NASA probe, designed to take exact measurements of gases in the atmosphere, is slated for launch in 2029 — followed by another mission that will take high-resolution images of the planet's surface. In conjunction with each other, these studies just might conclusively prove ancient tectonic activity on Venus — lending credence to the long-speculated hypothesis that the planet once harbored life, and may even have helped to seed the Earth with its building blocks.

Mike Floorwalker