In 1977, the New York Times published an article titled “Seeking an End to Cosmic Loneliness,” describing physicists’ attempts to pick up radio messages from aliens. The endeavor, known as the Search for Extraterrestrial Intelligence (SETI), was still in its early stages, and its proponents were struggling to persuade their peers and Congress that the idea was worth funding.
The quest to determine if anyone or anything is out there has gained greater scientific footing in the nearly half-century since that article’s publication. Back then, astronomers had yet to spot a single planet outside our solar system. Now we know the galaxy is teeming with a diversity of worlds. Our planet’s oceans were once considered exceptional, whereas evidence today suggests that numerous moons in the outer solar system host subsurface waters.
Our notion of the range of environments where life could exist has also expanded thanks to the discovery on Earth of extremophile organisms that can thrive in places far hotter, saltier, acidic, and more radioactive than previously thought possible, including creatures living around undersea hydrothermal vents.
We’re now getting closer than ever before to learning how common living worlds like ours actually are. New tools, including machine learning and artificial intelligence, could help scientists look past their preconceived notions of what constitutes life. Future instruments will sniff the atmospheres of distant planets and scan samples from our local solar system to see if they contain telltale chemicals in the right proportions for organisms to prosper.
“I think within our lifetime we will be able to do it,” says Ravi Kopparapu, a planetary scientist at NASA’s Goddard Space Flight Center in Maryland. “We will be able to know if there is life on other planets.”
While humans have a long history of speculating about distant worlds, for much of that time actual evidence was in short supply. The first planets around other stars—known as exoplanets—were discovered in the early 1990s, but it took until the launch of NASA’s Kepler space telescope in 2009 for astronomers to understand how common they were. Kepler carefully monitored hundreds of thousands of stars, looking for tiny dips in their brightness that could indicate planets passing in front of them. The mission helped the number of known exoplanets rise from a mere handful to over 5,500.
Kepler was built to help determine the prevalence of Earth-like planets orbiting sun-like stars at the right distance to have liquid water on their surface (a region often nicknamed the Goldilocks zone). While not a single extraterrestrial world has been a perfect twin of our own so far, researchers can use the sheer quantity of discoveries to make educated guesses as to how many might be out there. The current best estimates suggest that anywhere between 10% and 50% of sun-like stars have planets like ours, leading to numbers that make astronomers’ heads swim.
“If it’s 50%, that’s bonkers, right?” says Jessie Christiansen, an astrophysicist at Caltech in Pasadena, California. “There are billions of sun-like stars in the galaxy, and if half of them have Earth-like planets, there could be billions of habitable rocky planets.”
Is there anybody home?
Determining whether these planets actually contain organisms is no easy task. Researchers must capture the faint light from an exoplanet and spread it into its constituent wavelengths, scanning for signatures that indicate the presence and amount of different types of chemicals. While astronomers would like to focus on sun-like stars, doing so is technically challenging. NASA’s mighty new James Webb Space Telescope (JWST) is currently training its 6.5-meter mirror and unparalleled infrared instruments on worlds around stars smaller, cooler, and redder than our sun, known as M dwarfs. Such places might be habitable, but at the moment, nobody is really sure.
For liquid water to be present on their surfaces, planets around M dwarfs would need to orbit close to their stars—which tend to be more active than the sun, sending out violent flares that could strip away atmospheric gases and likely leave the ground a dry husk. JWST has been investigating Trappist-1, an M dwarf 40 light-years away with seven small rocky worlds, four of which are at the right distance to potentially have liquid water. The two closest exoplanets have already been shown to be devoid of atmospheres, but scientists are eagerly awaiting the results of JWST observations from the next three. They want to know if even those outside the habitable zone can have atmospheres.
There’s special interest in looking for other planets around M dwarf stars, because they are far more prevalent than sun-size stars. “If they find them to hold atmospheres, that increases the habitable real estate of the galaxy a hundredfold,” says Christiansen.
Once we’ve found a planet that looks a lot like Earth, then we’ll want to start hunting for chemical clues of life on its surface. JWST isn’t sensitive enough to do that, but future ground-based instruments like the Extremely Large Telescope, Giant Magellan Telescope, and Thirty Meter Telescope—which are expected to begin taking data in the 2030s—could tease out the chemical components of nearby Earth-like worlds. Information from more distant targets will have to wait for NASA’s next planned flagship mission, the space-based Habitable Worlds Observatory, expected to launch sometime in the late 2030s or early 2040s. The telescope will use either an external star shade or an instrument called a coronagraph to block the glaring light of a star and home in on dimmer planetary light and its potential molecular fingerprints.
Which chemicals in particular astronomers should be looking for remains a matter of debate. Ideally, they want to find what are known as biosignatures—molecules like water, methane, and carbon dioxide present in amounts similar to what we find on Earth. What that means in practice isn’t always clear, since our planet has gone through many periods when it contained life yet the quantities of different chemicals varied wildly.
“Do you want it to detect an Archaean Earth, like 2 or 3 billion years ago?” asks Kopparapu. “Or from the Neoproterozoic, where there was a snowball Earth? Or do you want to detect the current Earth, where there is a lot of free oxygen, ozone, water, and CO2?”
There was much excitement recently when JWST spotted dimethyl sulfide, a molecule that on our world is made only by living things, on an exoplanet nearly nine times Earth’s size located 120 light-years away. The results which have yet to be confirmed, highlight the trickiness of such methods. If dimethyl sulfide is truly present in the planet’s atmosphere, then starlight should also break it down to form ethane, a molecule that has yet to be seen. “No single gas is a biosignature,” says Kopparapu. “You need to see a combination of them.” Last year, he and others in the community published a report emphasizing that any particular finding must be placed in the context of its stellar and planetary environment, since there could be many results that seemingly point to life yet have alternative explanations.
Why counts as life?
This problem—how to definitively differentiate between life and non-life—is a perennial one, whether you’re looking at distant planets or even phenomena here on Earth. Researchers may soon receive help from algorithmic techniques that can tease out associations too complex for the human brain to fathom. In recent experiments, Robert Hazen and his colleagues took 134 living and non-living samples (including petroleum, carbon-rich meteorites, ancient fossils, and a wasp that flew into their lab), vaporized them, and spread out their chemical constituents. Roughly 500,000 different attributes were identified within each sample’s molecular makeup and run through a machine-learning program.
“When we look at those 500,000 attributes, there are patterns that are unique to living things and patterns unique to non-living things,” says Hazen, a mineralogist and astrobiologist at the Carnegie Institution for Science.
After the software was trained on 70% of the specimens, the technique was able to recognize with 90% accuracy which of the remaining samples had a biological origin. The device that is used to spread out the chemical components of the samples is around seven inches long, small enough to be sent on missions to nearby ocean worlds like Jupiter’s Europa or Saturn’s Enceladus. NASA’s Perseverance rover carried a similar instrument to Mars, so Hazen thinks his team’s machine-learning algorithm could be adapted to sift through its data and hunt for organisms past or present there. And because it relies on molecular relationships rather than detecting specific organic chemicals like DNA or amino acids, which may not be used in other biospheres, the method could allow scientists to look for life entirely unlike what we have on Earth.
Such machine-learning applications are also starting to find use in SETI, which has in recent years pivoted toward looking for a broader array of visible evidence for tool-using extraterrestrial species than before. Most in the field are on the lookout for such technosignatures, defined as “some remotely detectable signature of technology that we can characterize with astronomical instrumentation,” says Sofia Sheikh of the SETI Institute. This could be a radio signal, but other evidence could include things like optical laser pulses, giant space-based engineering projects, atmospheric pollution, or even artificial probes that make their way to our solar system.
At the Zwicky Transient Facility near San Diego, California, which continuously searches the entire night sky for brief flashes of light coming from unknown sources, engineers are teaching artificial intelligence how to identify features that would not be expected from natural phenomena. “It’s at that point that we can start asking questions,” says Ashish Mahabal, an astronomer and data scientist at Caltech. The answers to such questions could help reveal novel astronomical events or, just maybe, a star surrounded by enormous solar panels that feed an energy-intensive alien society.
SETI researchers hope that by using such tools, they can help overcome some of their anthropocentric biases. Most recognize that our expectations of otherworldly beings are constrained by our own experience. For example, the search for signs of massive alien solar panels is often “based on this assumption that there’s always going to be an exponential need for energy,” says Sheikh.
Because of all the avenues currently being explored, many scientists believe that answers to our questions about extraterrestrial life are not far off. Yet ultimately, the question of our cosmic loneliness is a philosophical one.
For most of humanity’s history, we didn’t believe ourselves to be alone. We filled the heavens with gods, monsters, and mythic creatures. It is only in the modern age that our species has started to worry about its place in the universe. But whether or not any other part of it harbors life, the cosmos is our home. We can choose to be lonely or to embrace the beauty and wonder all around us.
These robots know when to ask for help
A new training model, dubbed “KnowNo,” aims to address this problem by teaching robots to ask for our help when orders are unclear. At the same time, it ensures they seek clarification only when necessary, minimizing needless back-and-forth. The result is a smart assistant that tries to make sure it understands what you want without bothering you too much.
Andy Zeng, a research scientist at Google DeepMind who helped develop the new technique, says that while robots can be powerful in many specific scenarios, they are often bad at generalized tasks that require common sense.
For example, when asked to bring you a Coke, the robot needs to first understand that it needs to go into the kitchen, look for the refrigerator, and open the fridge door. Conventionally, these smaller substeps had to be manually programmed, because otherwise the robot would not know that people usually keep their drinks in the kitchen.
That’s something large language models (LLMs) could help to fix, because they have a lot of common-sense knowledge baked in, says Zeng.
Now when the robot is asked to bring a Coke, an LLM, which has a generalized understanding of the world, can generate a step-by-step guide for the robot to follow.
The problem with LLMs, though, is that there’s no way to guarantee that their instructions are possible for the robot to execute. Maybe the person doesn’t have a refrigerator in the kitchen, or the fridge door handle is broken. In these situations, robots need to ask humans for help.
KnowNo makes that possible by combining large language models with statistical tools that quantify confidence levels.
When given an ambiguous instruction like “Put the bowl in the microwave,” KnowNo first generates multiple possible next actions using the language model. Then it creates a confidence score predicting the likelihood that each potential choice is the best one.
The Download: inside the first CRISPR treatment, and smarter robots
The news: A new robot training model, dubbed “KnowNo,” aims to teach robots to ask for our help when orders are unclear. At the same time, it ensures they seek clarification only when necessary, minimizing needless back-and-forth. The result is a smart assistant that tries to make sure it understands what you want without bothering you too much.
Why it matters: While robots can be powerful in many specific scenarios, they are often bad at generalized tasks that require common sense. That’s something large language models could help to fix, because they have a lot of common-sense knowledge baked in. Read the full story.
Medical microrobots that travel inside the body are (still) on their way
The human body is a labyrinth of vessels and tubing, full of barriers that are difficult to break through. That poses a serious hurdle for doctors. Illness is often caused by problems that are hard to visualize and difficult to access. But imagine if we could deploy armies of tiny robots into the body to do the job for us. They could break up hard-to-reach clots, deliver drugs to even the most inaccessible tumors, and even help guide embryos toward implantation.
We’ve been hearing about the use of tiny robots in medicine for years, maybe even decades. And they’re still not here. But experts are adamant that medical microbots are finally coming, and that they could be a game changer for a number of serious diseases. Read the full story.
5 things we didn’t put on our 2024 list of 10 Breakthrough Technologies
We haven’t always been right (RIP, Baxter), but we’ve often been early to spot important areas of progress (we put natural-language processing on our very first list in 2001; today this technology underpins large language models and generative AI tools like ChatGPT).
Every year, our reporters and editors nominate technologies that they think deserve a spot, and we spend weeks debating which ones should make the cut. Here are some of the technologies we didn’t pick this time—and why we’ve left them off, for now.
New drugs for Alzheimer’s disease
Alzmeiher’s patients have long lacked treatment options. Several new drugs have now been proved to slow cognitive decline, albeit modestly, by clearing out harmful plaques in the brain. In July, the FDA approved Leqembi by Eisai and Biogen, and Eli Lilly’s donanemab could soon be next. But the drugs come with serious side effects, including brain swelling and bleeding, which can be fatal in some cases. Plus, they’re hard to administer—patients receive doses via an IV and must receive regular MRIs to check for brain swelling. These drawbacks gave us pause.
Sustainable aviation fuel
Alternative jet fuels made from cooking oil, leftover animal fats, or agricultural waste could reduce emissions from flying. They have been in development for years, and scientists are making steady progress, with several recent demonstration flights. But production and use will need to ramp up significantly for these fuels to make a meaningful climate impact. While they do look promising, there wasn’t a key moment or “breakthrough” that merited a spot for sustainable aviation fuels on this year’s list.
One way to counteract global warming could be to release particles into the stratosphere that reflect the sun’s energy and cool the planet. That idea is highly controversial within the scientific community, but a few researchers and companies have begun exploring whether it’s possible by launching a series of small-scale high-flying tests. One such launch prompted Mexico to ban solar geoengineering experiments earlier this year. It’s not really clear where geoengineering will go from here or whether these early efforts will stall out. Amid that uncertainty, we decided to hold off for now.