Science – Page 6

What ice fishing can teach us about making foraging decisions

cognitive psychology, decision making, ice fishing, psychology, Science, social sciences / Shannon Garcia / January 29, 2026

Ice fishing is a longstanding tradition in Nordic countries, with competitions proving especially popular. Those competitions can also tell scientists something about how social cues influence how we make foraging decisions, according to a new paper published in the journal Science.

Humans are natural foragers in even the most extreme habitats, digging up tubers in the tropics, gathering mushrooms, picking berries, hunting seals in the Arctic, and fishing to meet our dietary needs. Human foraging is sufficiently complex that scientists believe that meeting so many diverse challenges helped our species develop memory, navigational abilities, social learning skills, and similar advanced cognitive functions.

Researchers are interested in this question not just because it could help refine existing theories of social decision-making, but also could improve predictions about how different groups of humans might respond and adapt to changes in their environment. Per the authors, prior research in this area has tended to focus on solitary foragers operating in a social vacuum. And even when studying social foraging decisions, it’s typically done using computational modeling and/or in the laboratory.

“We wanted to get out of the lab,” said co-author Ralf Kurvers of Max Planck Institute for Human Development and TU Berlin. “The methods commonly used in cognitive psychology are difficult to scale to large, real-world social contexts. Instead, we took inspiration from studies of animal collective behavior, which routinely use cameras to automatically record behavior and GPS to provide continuous movement data for large groups of animals.”

Kurvers et al. organized 10 three-hour ice-fishing competitions on 10 lakes in eastern Finland for their study, with 74 experienced ice fishers participating. Each ice fisher wore a GPS tracker and a head-mounted camera so that the researchers could capture real-time data on their movements, interactions, and how successful they were in their fishing attempts. All told, they recorded over 16,000 individual decisions specifically about location choice and when to change locations. That data was then compared to the team’s computational cognitive models and agent-based simulations.

What ice fishing can teach us about making foraging decisions Read More »

New OpenAI tool renews fears that “AI slop” will overwhelm scientific research

Academic publishing, AI, AI assistants, ai slop, AI writing, Cambridge University Press, cornell university, GPT-5, Kevin Weil, large language models, machine learning, openai, peer review, Science, Science journal, scientific research, Yian Yin / Beth Washington / January 29, 2026

New “Prism” workspace launches just as studies show AI-assisted papers are flooding journals with diminished quality.

On Tuesday, OpenAI released a free AI-powered workspace for scientists. It’s called Prism, and it has drawn immediate skepticism from researchers who fear the tool will accelerate the already overwhelming flood of low-quality papers into scientific journals. The launch coincides with growing alarm among publishers about what many are calling “AI slop” in academic publishing.

To be clear, Prism is a writing and formatting tool, not a system for conducting research itself, though OpenAI’s broader pitch blurs that line.

Prism integrates OpenAI’s GPT-5.2 model into a LaTeX-based text editor (a standard used for typesetting documents), allowing researchers to draft papers, generate citations, create diagrams from whiteboard sketches, and collaborate with co-authors in real time. The tool is free for anyone with a ChatGPT account.

“I think 2026 will be for AI and science what 2025 was for AI in software engineering,” Kevin Weil, vice president of OpenAI for Science, told reporters at a press briefing attended by MIT Technology Review. He said that ChatGPT receives about 8.4 million messages per week on “hard science” topics, which he described as evidence that AI is transitioning from curiosity to core workflow for scientists.

OpenAI built Prism on technology from Crixet, a cloud-based LaTeX platform the company acquired in late 2025. The company envisions Prism helping researchers spend less time on tedious formatting tasks and more time on actual science. During a demonstration, an OpenAI employee showed how the software could automatically find and incorporate relevant scientific literature, then format the bibliography.

But AI models are tools, and any tool can be misused. The risk here is specific: By making it easy to produce polished, professional-looking manuscripts, tools like Prism could flood the peer review system with papers that don’t meaningfully advance their fields. The barrier to producing science-flavored text is dropping, but the capacity to evaluate that research has not kept pace.

When asked about the possibility of the AI model confabulating fake citations, Weil acknowledged in the press demo that “none of this absolves the scientist of the responsibility to verify that their references are correct.”

Unlike traditional reference management software (such as EndNote), which has formatted citations for over 30 years without inventing them, AI models can generate plausible-sounding sources that don’t exist. Weil added: “We’re conscious that as AI becomes more capable, there are concerns around volume, quality, and trust in the scientific community.”

The slop problem

Those concerns are not hypothetical, as we have previously covered. A December 2025 study published in the journal Science found that researchers using large language models to write papers increased their output by 30 to 50 percent, depending on the field. But those AI-assisted papers performed worse in peer review. Papers with complex language written without AI assistance were most likely to be accepted by journals, while papers with complex language likely written by AI models were less likely to be accepted. Reviewers apparently recognized that sophisticated prose was masking weak science.

“It is a very widespread pattern across different fields of science,” Yian Yin, an information science professor at Cornell University and one of the study’s authors, told the Cornell Chronicle. “There’s a big shift in our current ecosystem that warrants a very serious look, especially for those who make decisions about what science we should support and fund.”

Another analysis of 41 million papers published between 1980 and 2025 found that while AI-using scientists receive more citations and publish more papers, the collective scope of scientific exploration appears to be narrowing. Lisa Messeri, a sociocultural anthropologist at Yale University, told Science magazine that these findings should set off “loud alarm bells” for the research community.

“Science is nothing but a collective endeavor,” she said. “There needs to be some deep reckoning with what we do with a tool that benefits individuals but destroys science.”

Concerns about AI-generated scientific content are not new. In 2022, Meta pulled a demo of Galactica, a large language model designed to write scientific literature, after users discovered it could generate convincing nonsense on any topic, including a wiki entry about a fictional research paper called “The benefits of eating crushed glass.” Two years later, Tokyo-based Sakana AI announced “The AI Scientist,” an autonomous research system that critics on Hacker News dismissed as producing “garbage” papers. “As an editor of a journal, I would likely desk-reject them,” one commenter wrote at the time. “They contain very limited novel knowledge.”

The problem has only grown worse since then. In his first editorial of 2026 for Science, Editor-in-Chief H. Holden Thorp wrote that the journal is “less susceptible” to AI slop because of its size and human editorial investment, but he warned that “no system, human or artificial, can catch everything.” Science currently allows limited AI use for editing and gathering references but requires disclosure for anything beyond that and prohibits AI-generated figures.

Mandy Hill, managing director of academic publishing at Cambridge University Press & Assessment, has been even more blunt. In October 2025, she told Retraction Watch that the publishing ecosystem is under strain and called for “radical change.” She explained to the University of Cambridge publication Varsity that “too many journal articles are being published, and this is causing huge strain” and warned that AI “will exacerbate” the problem.

Accelerating science or overwhelming peer review?

OpenAI is serious about leaning on its ability to accelerate science, and the company laid out its case for AI-assisted research in a report published earlier this week. It profiles researchers who say AI models have sped up their work, including a mathematician who used GPT-5.2 to solve an open problem in optimization over three evenings and a physicist who watched the model reproduce symmetry calculations that had taken him months to derive.

Those examples go beyond writing assistance into using AI for actual research work, a distinction OpenAI’s marketing intentionally blurs. For scientists who don’t speak English fluently, AI writing tools could legitimately accelerate the publication of good research. But that benefit may be offset by a flood of mediocre submissions jamming up an already strained peer-review system.

Weil told MIT Technology Review that his goal is not to produce a single AI-generated discovery but rather “10,000 advances in science that maybe wouldn’t have happened or wouldn’t have happened as quickly.” He described this as “an incremental, compounding acceleration.”

Whether that acceleration produces more scientific knowledge or simply more scientific papers remains to be seen. Nikita Zhivotovskiy, a statistician at UC Berkeley not connected to OpenAI, told MIT Technology Review that GPT-5 has already become valuable in his own work for polishing text and catching mathematical typos, making “interaction with the scientific literature smoother.”

But by making papers look polished and professional regardless of their scientific merit, AI writing tools may help weak research clear the initial screening that editors and reviewers use to assess presentation quality. The risk is that conversational workflows obscure assumptions and blur accountability, and they might overwhelm the still very human peer review process required to vet it all.

OpenAI appears aware of this tension. Its public statements about Prism emphasize that the tool will not conduct research independently and that human scientists remain responsible for verification.

Still, one commenter on Hacker News captured the anxiety spreading through technical communities: “I’m scared that this type of thing is going to do to science journals what AI-generated bug reports is doing to bug bounties. We’re truly living in a post-scarcity society now, except that the thing we have an abundance of is garbage, and it’s drowning out everything of value.”

Benj Edwards is Ars Technica’s Senior AI Reporter and founder of the site’s dedicated AI beat in 2022. He’s also a tech historian with almost two decades of experience. In his free time, he writes and records music, collects vintage computers, and enjoys nature. He lives in Raleigh, NC.

New OpenAI tool renews fears that “AI slop” will overwhelm scientific research Read More »

Trade wars muzzle allied talks on Trump’s Golden Dome missile shield

golden dome, michael guetlein, military space, missile defense, Science, Space, space force / 9u50fv / January 27, 2026

“International partners, I have not been allowed to talk to yet because of the trade wars.”

President Donald Trump speaks in the Oval Office at the White House on May 20, 2025. Credit: Chip Somodevilla/Getty Images

Gen. Michael Guetlein, the senior officer in charge of the US military’s planned Golden Dome missile defense shield, has laid out an audacious schedule for deploying a network of space-based sensors and interceptors by the end of President Donald Trump’s term in the White House.

The three-year timeline is aggressive, with little margin for error in the event of budget or technological setbacks. The shield is designed to defend the US homeland against a range of long-range weapons, including intercontinental ballistic missiles (ICBMs), cruise missiles, and newer threats like hypersonic weapons and drones.

“By the summer of ’28, we will be able to defend the entire nation against ballistic missiles, as well as other generation aerial threats, and we will continue to grow that architecture through 2035,” Guetlein said Friday in a presentation to representatives from the US defense industry.

Supporters of Golden Dome say it is necessary to defend the United States against emerging threats from potential adversaries like Russia and China, each of which has a vast arsenal of ballistic missiles and more maneuverable hypersonic missiles that are difficult to detect and even harder to track and shoot down. Critics cite Golden Dome’s high cost and its potential to disrupt the global order, an eventuality they say would make the US homeland more prone to attack.

Guetlein’s team is moving fast in many areas. The Pentagon has inked deals with companies to develop prototypes for space-based missile interceptors, and Golden Dome is underpinned by billions of dollars of preexisting investment in sensor technology, reusable launchers, and mass-produced satellites.

These investments give Golden Dome a head start that former President Ronald Reagan’s similarly scoped Strategic Defense Initiative lacked in the 1980s. The initiative, nicknamed “Star Wars,” was drastically downsized as the Cold War wound down and the government struggled with ballooning costs and technical roadblocks.

Consequences to the chaos

But progress in at least one important area has stalled on the Golden Dome program. The Pentagon will need cooperation, if not tangible contributions, from allies to create the best possible version of the missile defense shield. Fundamentally, it comes down to geography. Places like the Canadian Arctic and Greenland are well-positioned to detect incoming missiles coming over the horizon from Russia or China. They might also be useful for hosting next-gen ground-based interceptors, which today are based in Alaska and California.

Guetlein said his interactions with allies have been limited. It’s the consequence of another of the Trump administration’s core policies: tariffs.

“International partners, I have not been allowed to talk to yet because of the trade wars,” Guetlein said. “The administration wants to get all that laying flat before I start having conversations with our allied partners. But everything we are doing is allied by design.”

Without detailed discussions with allies, the Pentagon’s Golden Dome appears to be one-sided.

“We are already planning on how to integrate our allied partners into our architecture,” Guetlein said. [We’re] already understanding what they can bring to the fight, whether that be capability, whether that be through FMS (foreign military sales), or whether that be territory that we need to access to put sensors, or what have you.”

Canada and Denmark, which count Greenland as an autonomous territory, are longtime allies of the United States. Both nations are members of NATO, and Canada is part of the North American Aerospace Defense Command (NORAD), an organization whose mission is already aligned with that of the Golden Dome. However, the Trump administration has targeted both nations for high tariffs, apparently in a bid to achieve the White House’s trade and security objectives.

Pituffik Space Base, formerly Thule Air Base, with the domes of the Thule Tracking Station, is pictured in northern Greenland on October 4, 2023. Credit: Thomas Traasdahl/Ritzau Scanpix/AFP via Getty Images

Trump’s obsession with acquiring Greenland, where the Space Force already has a military base, soured US relations with Denmark over the last year. The White House threatened additional tariffs on Denmark and other European countries until Denmark agreed to sell Greenland to the United States.

Then, after strong European pushback, Trump suddenly reversed course last week at the World Economic Forum in Davos, Switzerland. He withdrew threats to take Greenland by force and announced an unspecified “framework” with NATO for a future deal regarding the island. Trump told CNBC the framework involved partnerships on missile defense and mining: “They’re going to be involved in the Golden Dome, and they’re going to be involved in mineral rights, and so are we.”

In a statement, Danish Prime Minister Mette Frederiksen responded to Trump signaling an openness to a “constructive dialogue” on Arctic security, including the Golden Dome, “provided that this is done with respect for our territorial integrity.”

The White House’s backpedaling may have lowered the temperature of US relations with Denmark, at least for now. But Canada is still in Trump’s crosshairs. In a social media post on Friday, Trump claimed Canada was “against” plans to include Greenland in the Golden Dome program. Kirsten Hillman, Canada’s ambassador to the United States, told CBS News she didn’t know what Trump was talking about.

The back-and-forth followed Canadian Prime Minister Mark Carney’s rebuke of Trump’s foreign policy during a speech at Davos last week. In another sign of the withering US-Canada partnership, Trump threatened in a social media post Saturday to impose a 100 percent tariff on Canada if it follows through on a trade deal with China.

Canadian officials last year disclosed they were in talks to participate in the Golden Dome. In an interview with The War Zone published earlier this month, the operational commander of NORAD’s Canadian region advocated for “integrated missile defense” and “ground-based effectors” but said Canada’s role in a space-based layer would be a “political decision.”

“We look at it as Continental Shield,” said Royal Canadian Air Force Maj. Gen. Chris McKenna. “Golden Dome is the US brand on it. From our point of view, it’s great air missile defense and what we will put on the table to defend the continent with. And so I think there are ongoing negotiations between our governments with respect to what the specific investments will be.”

A roadmap to something

Guetlein’s comments on Golden Dome last week were part of an “industry day” hosted by Space Systems Command, the Los Angeles-based unit charged with procurement for most of the Space Force’s major programs. Asked what his top priority is this year, Guetlein said his emphasis is on the shield’s command and control system, the “glue layer” that ties together everything else necessary to make Golden Dome work. His deadline to demonstrate the command-and-control system is this summer.

The kinetic, or hit-to-kill, elements of the missile shield will come next. “In the summer of ’27, I have to integrate the interceptors, and in ’28 I have to demonstrate operational capability that is fielded against credible threats,” Guetlein said.

This would bring some version of Golden Dome into service before President Trump leaves the White House, a timetable used for several other programs, including NASA’s Artemis mission to land astronauts on the Moon. Officials are willing to trade perfection for speed. Guetlein’s intention is to “rapidly get an 85 percent product in the field.”

Gen. Michael Guetlein, overseeing the development of the Golden Dome missile defense system, looks on as President Donald Trump speaks in the Oval Office of the White House on May 20, 2025, in Washington, DC. Credit: Jim Watson/AFP via Getty Images

But a comprehensive missile defense shield, if economically and technically feasible, would take many more years to complete. Some pieces of what could become Golden Dome already exist, such as ground-based midcourse interceptors and an array of missile detection sensors on land, at sea, and in space.

Other elements were well along in development before Trump’s second term in the White House. These include a constellation of low-Earth-orbit satellites to better track hypersonic missiles and relay data to help US and allied forces shoot them down. This network, led by the Space Development Agency, began launching in 2023 but is not yet operational.

The Pentagon has already kicked off the preliminary steps to buy space-based interceptors. The shield will need hundreds or thousands of orbiting platforms on alert to shoot down a ballistic missile in flight. The Defense Department also aims to field modernized ground-based interceptors and Golden Dome’s command and control network, the system Guetlein identified as his top priority for this year.

Satellites are getting cheaper and easier to build. SpaceX has manufactured and launched approximately 11,000 Starlink satellites in less than seven years. No other company comes close to this number, but there are businesses looking at building standardized satellites that could be integrated with a range of payloads, including Earth-viewing cameras, radars, communication antennas, or perhaps, a missile interceptor.

But the interceptors themselves are brand new. Nothing like them has ever been put into space and tested before. Last year, the Space Force asked companies to submit their ideas for space-based interceptor (SBI) prototypes. The military awarded 18 companies small prototype development deals in November. The Space Force did not identify the winners, and the value of each contract fell below the $9 million public disclosure threshold for Pentagon programs.

These 18 awards are focused on boost-phase SBIs, weapons that can take out a missile soon after it leaves its silo. From a physics perspective, this is one of the most difficult things to ask an interceptor to do, because the weapon must account for atmospheric disturbances and reentry heating to reach its target. In December, the Space Force issued a follow-up request for prototype proposals looking at space-based midcourse interceptors capable of destroying a ballistic missile while it is coasting through space.

The Pentagon has not said how many SBIs it will need for Golden Dome, or what they will look like. Essentially, an orbiting interceptor will be a flying fuel tank with a rocket and a sensor package to home in on its target. But the Space Force and its prospective Golden Dome contractors, which include industry giants Northrop Grumman and Lockheed Martin, have not disclosed the interceptors’ design specs or how many the shield will need.

A Standard Missile 3 Block IIA launches from the Aegis Ashore Missile Defense Test Complex at the Pacific Missile Range Facility in Kauai, Hawaii, on December 10, 2018, during a test to intercept an intermediate-range ballistic missile target in space. Credit: Mark Wright/DOD

“I can move money around at will”

Guetlein said Pentagon leaders have empowered him with “unprecedented authorities” on the Golden Dome program. He said Friday his office has full authority over the program’s technical aspects, along with Golden Dome’s procurement, contracting, hiring, security, and budget.

President Trump directed the military to start the Golden Dome program in an executive order last January and nominated Guetlein to head the effort. The Senate confirmed him for the job in a voice vote last July. After the initial pomp of Golden Dome’s announcement, including multiple Oval Office photo ops, military officials went quiet. Guetlein cited security concerns for the Pentagon’s silence on one of Trump’s marquee defense programs.

“I was confirmed on the 18th of July. On the 19th of July, I popped No. 1 on the intel collect list for our adversaries,” Guetlein said. “On the 20th of July, they started hacking our defense industrial base, and the Secretary (of Defense) asked us to go silent. So, we have been quiet. I have not been talking to industry consortiums. I have not been talking to the press. I have not been talking to the think tanks. And it wasn’t until September that I was allowed to even start talking to the Hill (Congress).”

Despite this absence of dialogue, lawmakers approved $23 billion in a down payment for Golden Dome last summer. The final cost of the system is anyone’s guess, but the deployment and sustainment of Golden Dome will certainly cost hundreds of billions of dollars over multiple decades. Just getting the missile shield through a partial activation will cost $175 billion over three years, according to the White House.

Guetlein is off and running with the one-time infusion of $23 billion Congress gave him last year. “The budget authority… for the first $23 billion, it has no strings, no color, no year,” he said. “I can move that money around at will.”

He’s about to receive another $13.4 billion in this year’s defense budget, pending the appropriation bill’s final passage in the Senate and President Trump’s signature.

“I’ve got the Space Force buying SBIs. I’ve got the Army buying munitions and sensors. I’ve got the Navy buying munitions. I’ve got the Missile Defense Agency buying next-generation interceptors, glide phase interceptors, and a whole host of other capabilities,” Guetlein said.

Despite their willingness to fund Golden Dome, lawmakers still aren’t satisfied with the information they are getting from the Pentagon. In a document adjoining this year’s defense budget bill, members of the House and Senate defense appropriations subcommittees wrote they were “unable to effectively assess resources available” and unable to “conduct oversight” of the Golden Dome program “due to insufficient budgetary information” provided by the Department of Defense.

The lawmakers wrote that they “support the operational objectives of Golden Dome” and directed the Pentagon to “provide a comprehensive spend plan” for the missile shield within 60 days of the budget bill becoming law.

“Unfortunately because of the intel threat, we can only brief you at the classified level on what the architecture is,” Guetlein said.

Stephen Clark is a space reporter at Ars Technica, covering private space companies and the world’s space agencies. Stephen writes about the nexus of technology, science, policy, and business on and off the planet.

Trade wars muzzle allied talks on Trump’s Golden Dome missile shield Read More »

Doctors face-palm as RFK Jr.’s top vaccine advisor questions need for polio shot

ACIP, anti-vaccine, CDC, Kirk Milhoan, measles, polio, robert f kennedy jr, Science, vaccines / DJ Henderson / January 26, 2026

He then pondered out loud what would happen if people stopped getting vaccinated. “If we take away all of the herd immunity, then does that switch, does that teeter-totter switch in a different direction?” he asked.

Backlash

In a statement, AMA Trustee Sandra Adamson Fryhofer blasted the question. “This is not a theoretical debate—it is a dangerous step backward,” she said. “Vaccines have saved millions of lives and virtually eliminated devastating diseases like polio in the United States. There is no cure for polio. When vaccination rates fall, paralysis, lifelong disability, and death return. The science on this is settled.”

Fryhofer also took aim at Milhoan’s repeated argument that the focus of vaccination policy should move from population-level health to individual autonomy. Moving away from routine immunizations, which include discussions between clinicians and patients, “does not increase freedom—it increases suffering,” she said, adding that the weakening of recommendations “will cost lives.”

Overall, Milhoan’s comments only further erode the relevance of ACIP and federal vaccine policy among the medical community and states. According to a KFF policy brief, 27 states and Washington, DC, have already announced they will not follow current CDC vaccine recommendations, which Kennedy dramatically overhauled earlier this month without even consulting the ACIP. Instead, the majority of states are relying on previous recommendations or recommendations made within states or by medical organizations.

On Monday, the American Academy of Pediatrics announced the 2026 update to its childhood and adolescent vaccine schedule, which it has held up as an alternative to the CDC’s schedule and has been widely embraced by pediatricians. In the announcement, AAP noted that 12 other medical organizations have endorsed the schedule, including the AMA, the American Academy of Family Physicians, the American College of Obstetricians and Gynecologists, the Infectious Diseases Society of America, and the Pediatric Infectious Diseases Society.

The AAP’s updated recommendations are largely the same as the schedule from last year, but it is significantly different from the CDC’s recommendations, which “depart from longstanding medical evidence and no longer offer the optimal way to prevent illnesses in children,” the AAP said.

“The AAP will continue to provide recommendations for immunizations that are rooted in science and are in the best interest of the health of infants, children and adolescents of this country,” AAP President Andrew Racine said in the announcement.

Doctors face-palm as RFK Jr.’s top vaccine advisor questions need for polio shot Read More »

Former astronaut on lunar spacesuits: “I don’t think they’re great right now”

artemis, axiom space, Commercial space, Features, human spaceflight, kate rubins, mike barratt, NASA, Science, Space, spacesuits / Mike M. / January 26, 2026

“These are just the difficulties of designing a spacesuit for the lunar environment.”

NASA astronaut Loral O’Hara kneels down to pick up a rock during testing of Axiom’s lunar spacesuit inside NASA’s Neutral Buoyancy Laboratory in Houston on September 24, 2025. Credit: NASA

Crew members traveling to the lunar surface on NASA’s Artemis missions should be gearing up for a grind. They will wear heavier spacesuits than those worn by the Apollo astronauts, and NASA will ask them to do more than the first Moonwalkers did more than 50 years ago.

The Moonwalking experience will amount to an “extreme physical event” for crews selected for the Artemis program’s first lunar landings, a former NASA astronaut told a panel of researchers, physicians, and engineers convened by the National Academies.

Kate Rubins, who retired from the space agency last year, presented the committee with her views on the health risks for astronauts on lunar missions. She outlined the concerns NASA officials often talk about: radiation exposure, muscle and bone atrophy, reduced cardiovascular and immune function, and other adverse medical effects of spaceflight.

Scientists and astronauts have come to understand many of these effects after a quarter-century of continuous human presence on the International Space Station. But the Moon is different in a few important ways. The Moon is outside the protection of the Earth’s magnetosphere, lunar dust is pervasive, and the Moon has partial gravity, about one-sixth as strong as the pull we feel on Earth.

Each of these presents challenges for astronauts living and working on the lunar surface, and their effects are amplified for crew members who venture outside for spacewalks. NASA selected Axiom Space, a Houston-based company, for a $228 million fixed-price contract to develop commercial pressurized spacesuits for the Artemis III mission, slated to be the first human landing mission on the Moon since 1972.

NASA hopes to fly the Artemis III mission by the end of 2028, but the schedule is in question. The readiness of Axiom’s spacesuits and the availability of new human-rated landers from SpaceX and Blue Origin are driving the timeline for Artemis III.

Stressing about stress

Rubins is a veteran of two long-duration spaceflights on the International Space Station, logging 300 days in space and conducting four spacewalks totaling nearly 27 hours. She is also an accomplished microbiologist and became the first person to sequence DNA in space.

“What I think we have on the Moon that we don’t really have on the space station that I want people to recognize is an extreme physical stress,” Rubins said. “On the space station, most of the time you’re floating around. You’re pretty happy. It’s very relaxed. You can do exercise. Every now and then, you do an EVA (Extravehicular Activity, or spacewalk).”

“When we get to the lunar surface, people are going to be sleep shifting,” Rubins said. “They’re barely going to get any sleep. They’re going to be in these suits for eight or nine hours. They’re going to be doing EVAs every day. The EVAs that I did on my flights, it was like doing a marathon and then doing another marathon when you were done.”

NASA astronaut Kate Rubins inside the International Space Station in 2020. Credit: NASA

Rubins is now a professor of computational and systems biology at the University of Pittsburgh School of Medicine. She said treks on the Moon will be “even more challenging” than her spacewalks outside the ISS.

The Axiom spacesuit design builds on NASA’s own work developing a prototype suit to replace the agency’s decades-old Extravehicular Mobility Units (EMUs) used for spacewalks at the International Space Station (ISS). The new suits allow for greater mobility, with more flexible joints to help astronauts use their legs, crouch, and bend down—things they don’t have to do when floating outside the ISS.

Astronauts on the Moon also must contend with gravity. Including a life-support backpack, the commercial suit weighs more than 300 pounds in Earth’s gravity, but Axiom considers the exact number proprietary. The Axiom suit is considerably heavier than the 185-pound spacesuit the Apollo astronauts wore on the Moon. NASA’s earlier prototype exploration spacesuit was estimated to weigh more than 400 pounds, according to a 2021 report by NASA’s inspector general.

“We’ve definitely seen trauma from the suits, from the actual EVA suit accommodation,” said Mike Barratt, a NASA astronaut and medical doctor. “That’s everything from skin abrasions to joint pain to—no kidding—orthopedic trauma. You can potentially get a fracture of sorts. EVAs on the lunar surface with a heavily loaded suit and heavy loads that you’re either carrying or tools that you’re reacting against, that’s an issue.”

On paper, the Axiom suits for NASA’s Artemis missions are more capable than the Apollo suits. They can support longer spacewalks and provide greater redundancy, and they’re made of modern materials to enhance flexibility and crew comfort. But the new suits are heavier, and for astronauts used to spacewalks outside the ISS, walks on the Moon will be a slog, Rubins said.

“I think the suits are better than Apollo, but I don’t think they are great right now,” Rubins said. “They still have a lot of flexibility issues. Bending down to pick up rocks is hard. The center of gravity is an issue. People are going to be falling over. I think when we say these suits aren’t bad, it’s because the suits have been so horrible that when we get something slightly less than horrible, we get all excited and we celebrate.”

The heavier lunar suits developed for Artemis missions run counter to advice from former astronaut Harrison “Jack” Schmitt, who spent 22 hours walking on the Moon during NASA’s Apollo 17 mission in 1972.

“I’d have that go about four times the mobility, at least four times the mobility, and half the weight,” Schmitt said in a NASA oral history interview in 2000. “Now, one way you can… reduce the weight is carry less consumables and learn to use consumables that you have in some other vehicle, like a lunar rover. Any time you’re on the rover, you hook into those consumables and live off of those, and then when you get off, you live off of what’s in your backpack. We, of course, just had the consumables in our backpack.”

NASA won’t have a rover on the first Artemis landing mission. That will come on a later flight. A fully pressurized vehicle for astronauts to drive across the Moon may be ready sometime in the 2030s. Until then, Moonwalkers will have to tough it out.

“I do crossfit. I do triathlons. I do marathons. I get out of a session in the pool in the NBL (Neutral Buoyancy Laboratory) doing the lunar suit underwater, and I just want to go home and take a nap,” Rubins told the panel. “I am absolutely spent. You’re bruised. This is an extreme physical event in a way that the space station is not.”

NASA astronaut Mike Barratt inside the International Space Station in 2024. Credit: NASA

Barratt met with the same National Academies panel this week and presented a few hours before Rubins. The committee was chartered to examine how human explorers can enable scientific discovery at sites across the lunar surface. Barratt had a more favorable take on the spacesuit situation.

“This is not a commercial for Axiom. I don’t promote anyone, but their suit is getting there,” Barratt said. “We’ve got 700 hours of pressurized experience in it right now. We do a lot of tests in the NBL, and there are techniques and body conditioning that you do to help you get ready for doing things like this. Bending down in the suit is really not too bad at all.”

Rubins and Barratt did not discuss the schedule for when Axiom’s lunar spacesuit will be ready to fly to the Moon, but the conversation illuminated the innumerable struggles of spacewalking, Moonwalking, and the training astronauts undergo to prepare for extravehicular outings.

The one who should know

I spoke directly with Rubins after her discussion with the National Academies. Her last assignment at NASA was as chief of the EVA and robotics branch in the astronaut office, where she assisted in the development of the new lunar spacesuits. I asked about her experiences testing the lunar suit and her thoughts on how astronauts should prepare for Moonwalks.

“The suits that we have are definitely much better than Apollo,” Rubins said in the interview. “They were just big bags of air. The joints aren’t in there, so it was harder to move. What they did have going for them was that they were much, much lighter than our current spacesuits. We have added a lot of the joints back, and that does get some mobility for us. But at the end of the day, the suits are still quite heavy.”

You can divide the weight of the suit by six to get an idea of how it might feel to carry it around on the lunar surface. While it won’t feel like 300 pounds, astronauts will still have to account for their mass and momentum.

Rubins explained:

Instead of kind of floating in microgravity and moving your mass around with your hands and your arms, now we’re ambulating. We’re walking with our legs. You’re going to have more strain on your knees and your hips. Your hamstrings, your calves, and your glutes are going to come more into play.

I think, overall, it may be a better fit for humans physically because if you ask somebody to do a task, I’m going to be much better at a task if I can use my legs and I’m ambulating. Then I have to pull myself along with my arms… We’re not really built to do that, but we are built to run and to go long distances. Our legs are just such a powerful force.

So I think there are a lot of things lining up that are going to make the physiology easier. Then there are things that are going to be different because we’re now in a partial gravity environment. We’re going to be bending, we’re going to be twisting, we’re going to be doing different things.

It’s an incredibly hard engineering challenge. You have to keep a human alive in absolute vacuum, warm at temperatures that you know in the polar regions could go as far down as 40 Kelvin (minus 388° Fahrenheit). We haven’t sent humans anywhere that cold before. They are also going to be very hot. They’re going to be baking in the sunshine. You’ve got radiation. If you put all that together, that’s a huge amount of suit material just to keep the human physiology and the human body intact.

Then our challenge is ‘how do you make that mobile?’ It’s very difficult to bend down and pick up a rock. You have to manage that center of gravity because you’re wearing that big life support system on your back, a big pack that has a lot of mass in it, so that brings your center of gravity higher than you’re used to on Earth and a little bit farther backward.

When you move around, it’s like wearing a really, really heavy backpack that has mass but no weight, so it’s going to kind of tip you back. You can do some things with putting weights on the front of the suit to try to move that center of gravity forward, but it’s still higher, and it’s not exactly at your center of mass that you’re used to on the Earth. On the Earth, we have a center of our mass related to gravity, and nobody ever thinks about it, and you don’t think about it until it moves somewhere else, and then it makes all of your natural motion seem very difficult.

Those are some of the challenges that we’re facing engineering-wise. I think the new suits, they’ve gone a long way toward addressing these, but it’s still a hard engineering challenge. And I’m not talking about any specific suit. I can’t talk about the details of the provider’s suits. This is the NASA xEMU and all the lunar suits I have tested over the years. That includes the Mark III suit, the Axiom suit. They have similar issues. So this isn’t really anything about a specific vendor. These are just the difficulties of designing a spacesuit for the lunar environment.

NASA trains astronauts for spacewalks in the Neutral Buoyancy Laboratory, an enormous pool in Houston used for simulating weightlessness. They also use a gravity-offloading device to rehearse the basics of spacewalking. The optimal test environment, short of the space environment itself, will be aboard parabolic flights, where suit developers and astronauts can get the best feel for the suit’s momentum, according to Rubins.

Axiom and NASA are well along assessing the new lunar spacesuit’s performance underwater, but they haven’t put it through reduced-gravity flight testing. “Until you get to the actual parabolic flight, that’s when you can really test the ability to manage this momentum,” Rubins said.

NASA astronauts Loral O’Hara and Stan Love test Axiom’s lunar spacesuit inside NASA’s Neutral Buoyancy Laboratory in Houston on September 24, 2025. Credit: NASA

Recovering from a fall on the lunar surface comes with its own perils.

“You’re face down on the lunar surface, and you have to do the most massive, powerful push up to launch you and the entire mass of the suit up off the surface, high enough so you can then flip your legs under you and catch the ground,” Rubins said. “You basically have to kind of do a jumping pushup… This is a risky maneuver we test a whole bunch in training. It’s really non-trivial.”

The lunar suits are sleeker than the suits NASA uses on the ISS, but they are still bulky. “If you’re trying to kneel, if you’re thinking about bending forward at your waist, all that material in your waist has nowhere to go, so it just compresses and compresses,” Rubins said. “That’s why I say it’s harder to kneel. It’s harder to bend forward because you’re having to compress the suit in those areas.

“We’ve done these amazing things with joint mobility,” Rubins said. “The mobility around the joints is amazing… but now we’re dealing with this compression issue. And there’s not an obvious engineering fix to that.”

The fix to this problem might come in the form of tools instead of changes to the spacesuit itself. Rubins said astronauts could use a staff, or something like a hiking pole, to brace themselves when they need to kneel or bend down. “That way I’m not trying to compress the suit and deal with my balance at the same time.”

A bruising exertion

The Moonwalker suit can comfortably accommodate a wider range of astronauts than NASA’s existing EMUs on the space station. The old EMUs can be resized to medium, large, and extra-large, but that leaves gaps and makes the experience uncomfortable for a smaller astronaut. This discomfort is especially noticeable while practicing for spacewalks underwater, where the tug of gravity is still present, Rubins said.

“As a female, I never really had an EMU that fit me,” Rubins said. “It was always giant. When I’m translating around or doing something, I’m physically falling and slamming myself, my chest or my back, into one side of the suit or the other underwater, whereas with the lunar suit, I’ve got a suit that fits me right. That’s going to lead to less bruising. Just having a suit that fits you is much better.”

Mission planners should also emphasize physical conditioning for astronauts assigned to lunar landing missions. That includes preflight weight and endurance training, plus guidance on what to eat in space to maximize energy levels before astronauts head outside for a stroll.

“That human has to go up really maximally conditioned,” Rubins said.

Rubins and Barratt agreed that NASA and its spacesuit provider should be ready to rapidly respond to feedback from future Moonwalkers. Engineers modified and upgraded the Apollo spacesuits in a matter of months, iterating the design between each mission.

“Our general design is on a good path,” Rubins said. “We need to make sure that we continue to push for increasing improvements in human performance, and some of that ties back to the budget. Our first suit design is not where we’re going to be done if we want to do a really sustained lunar program. We have to continue to improve, and I think it’s important to recognize that we’re going to learn so many lessons during Artemis III.”

Barratt has a unique perspective on spacesuit design. He has performed spacewalks at the ISS in NASA’s spacesuit and the Russian Orlan spacesuit. Barratt said the US suit is easier to work in than the Orlan, but the Russian suit is “incredibly reliable” and “incredibly serviceable.”

“It had a couple of glitches, and literally, you unzip a curtain and it’s like looking at my old Chevy Blazer,” Barratt said. “Everything is right there. It’s mechanical, it’s accessible with standard tools. We can fix it. We can do that really easily. We’ve tried to incorporate those lessons learned into our next-generation EVA systems.”

Contrast that with the NASA suits on the ISS, where one of Barratt’s spacewalks in 2024 was cut short by a spacesuit water leak. “We recently had to return a suit from the space station,” Barratt said. “We’ve got another one that’s sort of offline for a while; we’re troubleshooting it. It’s a really subtle problem that’s extremely difficult to work on in places that are hard to access.”

It’s happened before. Apollo 17 astronaut Harrison “Jack” Schmitt loses his balance on the Moon, then quickly recovers. Credit: NASA

Harrison Schmitt, speaking with a NASA interviewer in 2000, said his productivity in the Apollo suit “couldn’t have been much more than 10 percent of what you would do normally here on Earth.”

“You take the human brain, the human eyes, and the human hands into space. That’s the only justification you have for having human beings in space,” Schmitt said. “It’s a massive justification, but that’s what you want to use, and all three have distinct benefits in productivity and in gathering new information and infusing data over any automated system. Unfortunately, we have discarded one of those, and that is the hands.”

Schmitt singled out the gloves as the “biggest problem” with the Apollo suits. “The gloves are balloons, and they’re made to fit,” he said. Picking something up with a firm grip requires squeezing against the pressure inside the suit. The gloves can also damage astronauts’ fingernails.

“That squeezing against that pressure causes these forearm muscles to fatigue very rapidly,” Schmitt said. “Just imagine squeezing a tennis ball continuously for eight hours or 10 hours, and that’s what you’re talking about.”

Barratt recounted a conversation in which Schmitt, now 90, said he wouldn’t have wanted to do another spacewalk after his three excursions with commander Gene Cernan on Apollo 17.

“Physically, and from a suit-maintenance standpoint, he thought that that was probably the limit, what they did,” Barratt said. “They were embedded with dust. The visors were abraded. Every time they brushed the dust off the visors, they lost visibility.”

Getting the Artemis spacesuit right is vital to the program’s success. You don’t want to travel all the way to the Moon and stop exploring because of sore fingers or an injured knee.

“If you look at what we’re spending on suits versus what we’re spending on the rocket, this is a pretty small amount,” Rubins said. “Obviously, the rocket can kill you very quickly. That needs to be done right. But the continuous improvement in the suit will get us that much more efficiency. Saving 30 minutes or an hour on the Moon, that gives you that much more science.”

“Once you have safely landed on the lunar surface, this is where you’ve got to put your money,” Barratt said.

Former astronaut on lunar spacesuits: “I don’t think they’re great right now” Read More »

Did Edison accidentally make graphene in 1879?

graphene, lightbulbs, materials science, Physics, Science, Thomas Edison / DJ Henderson / January 24, 2026

Graphene is the thinnest material yet known, composed of a single layer of carbon atoms arranged in a hexagonal lattice. That structure gives it many unusual properties that hold great promise for real-world applications: batteries, super capacitors, antennas, water filters, transistors, solar cells, and touchscreens, just to name a few. The physicists who first synthesized graphene in the lab won the 2010 Nobel Prize in Physics. But 19th century inventor Thomas Edison may have unknowingly created graphene as a byproduct of his original experiments on incandescent bulbs over a century earlier, according to a new paper published in the journal ACS Nano.

“To reproduce what Thomas Edison did, with the tools and knowledge we have now, is very exciting,” said co-author James Tour, a chemist at Rice University. “Finding that he could have produced graphene inspires curiosity about what other information lies buried in historical experiments. What questions would our scientific forefathers ask if they could join us in the lab today? What questions can we answer when we revisit their work through a modern lens?”

Edison didn’t invent the concept of incandescent lamps; there were several versions predating his efforts. However, they generally had a a very short life span and required high electric current, so they weren’t well suited to Edison’s vision of large-scale commercialization. He experimented with different filament materials starting with carbonized cardboard and compressed lampblack. This, too, quickly burnt out, as did filaments made with various grasses and canes, like hemp and palmetto. Eventually Edison discovered that carbonized bamboo made for the best filament, with life spans over 1200 hours using a 110 volt power source.

Lucas Eddy, Tour’s grad student at Rice, was trying to figure out ways to mass produce graphene using the smallest, easiest equipment he could manage, with materials that were both affordable and readily available. He considered such options as arc welders and natural phenomena like lightning striking trees—both of which he admitted were “complete dead ends.” Edison’s light bulb, Eddy decided, would be ideal, since unlike other early light bulbs, Edison’s version was able to achieve the critical 2000 degree C temperatures required for flash Joule heating—the best method for making so-called turbostratic graphene.

Did Edison accidentally make graphene in 1879? Read More »

This 67,800-year-old hand stencil is the world’s oldest human-made art

anthropology, Archaeology, australia, cave art, hominins, human migration, Indonesia, rock art, Sahul, Science / Beth Washington / January 24, 2026

generative AI could never

The world’s oldest art has an unintentional story to tell about human exploration.

These 17,000-year-old hand stencils from Liang Jarie Maros, in another area of Sulawesi, bear a striking resemblance to the much older ones in Liang Metanduno. Credit: OKtaviana et al. 2026

The world’s oldest surviving rock art is a faded outline of a hand on an Indonesian cave wall, left 67,800 years ago.

On a tiny island just off the coast of Sulawesi (a much larger island in Indonesia), a cave wall bears the stenciled outline of a person’s hand—and it’s at least 67,800 years old, according to a recent study. The hand stencil is now the world’s oldest work of art (at least until archaeologists find something even older), as well as the oldest evidence of our species on any of the islands that stretch between continental Asia and Australia.

Photo of an archaeologists examining a hand stencil painted on a cave wall, using a flashlight — Adhi Oktaviana examines a slightly more recent hand stencil on the wall of Liang Metanduno. Credit: Oktaviana et al. 2026

Hands reaching out from the past

Archaeologist Adhi Agus Oktaviana, of Indonesia’s National Research and Innovation Agency, and his colleagues have spent the last six years surveying 44 rock art sites, mostly caves, on Sulawesi’s southeastern peninsula and the handful of tiny “satellite islands” off its coast. They found 14 previously undocumented sites and used rock formations to date 11 individual pieces of rock art in eight caves—including the oldest human artwork discovered so far.

About 67,800 years ago, someone stood in the darkness of Liang Metanduno and placed their hand flat against the limestone wall. They, or maybe a friend, then blew a mixture of pigment and water onto the wall, covering and surrounding their hand. When they pulled their hand carefully away from the rock, careful not to disturb the still-wet paint, they left behind a crisp outline of their palm and fingers, haloed by a cloud of deep red.

The result is basically the negative of a handprint, and it’s a visceral, tangible link to the past. Someone once laid their hand on the cave wall right here, and you can still see its outline like a lingering ghost, reaching out from the other side of the rock. If you weren’t worried about damaging the already faded and fragile image, you could lay your hand in the same spot and meet them halfway.

Today, the stencil is so faded that you can barely see it, but if you look closely, it’s there: a faint halo of reddish-orange pigment, outlining the top part of a palm and the base of the fingers. A thin, nearly transparent layer of calcite covers the faded shape, left behind by millennia of water dripping down the cave wall. The ratio of uranium and thorium in a sheet of calcite suggests that it formed at least 71,000 years ago—so the outline of the hand beneath it must have been left behind sometime before that, probably around 67,800 years ago.

A photo of two figures on a cave wall, with the faint outline of a hand circled in black — The hand stencil is faded and overlain by more recent (but still ancient) artwork; it’s circled in black to help you find it in this photo. Credit: Oktaviana et al. 2026

That makes Liang Metanduno the home of the oldest known artwork in the world, beating the previous contender (a Neanderthal hand stencil in Spain) by about 1,100 years.

“These findings support the growing view that Sulawesi was host to a vibrant and longstanding artistic culture during the late Pleistocene epoch,” wrote Oktaviana and his colleagues in their recent paper.

The karst caves of Sulawesi’s southwestern peninsula, Maros-Pangkep, are a treasure trove of deeply ancient artwork: hand stencils, as well as drawings of wild animals, people, and strange figures that seem to blend the two. A cave wall at Liang Bulu’Sipong 4 features a 4.5-meter-long mural of humanlike figures facing off against wild pigs and dwarf buffalo, and a 2024 study pushed the mural’s age back to 51,200 years ago, making it the second-oldest artwork that we know of (after the Liang Metanduno hand stencil in the recent study).

Archaeologists have only begun to rediscover the rock art of Maros-Pangkep in the last decade or so, and other areas of the island, like Southeast Sulawesi and its tiny satellite islands, have received even less attention—so we don’t know what’s still there waiting for humanity to find again after dozens of millennia. We also don’t know what the ancient artist was trying to convey with the outline of their hand on the cave wall, but part of the message rings loud and clear across tens of millennia: At least 67,800 years ago, someone was here.

Really, really ancient mariners

The hand stencil on the wall of Liang Metanduno is, so far, the oldest evidence of our presence in Wallacea, the group of islands stretched between the continental shelves of Asia and Australia. Populating these islands is “widely considered to have involved the first planned, long-distance sea crossing undertaken by our species,” wrote Oktaviana and his colleagues.

Back when the long-lost artist laid their hand on the wall, sea levels were about 100 meters lower than they are today. Mainland Asia, Sumatra, and Borneo would have been high points in a single landmass, joined by wide swaths of lowlands that today lie beneath shallow ocean. The eastern shore of Borneo would have been a jumping-off point, beyond which lay several dozen kilometers of water and (out of view over the horizon) Sulawesi.

The first few people may have washed ashore on Sulawesi on some misadventure: lost fishermen or tsunami survivors, maybe. But at some point, people must have started making the crossing on purpose, which implies that they knew how to build rafts or boats, how to steer them, and that land awaited them on the other side.

Liang Metanduno pushes back the timing of that crossing by nearly 10,000 years. It also lends strong support to arguments that people arrived in Australia earlier than archaeologists had previously suspected. Archaeological evidence from a rock shelter called Madjedbebe, in northern Australia, suggests that people were living there by 65,000 years ago. But that evidence is still debated (such is the nature of archaeology), and some archaeologists argue that humans didn’t reach the continent until around 50,000 years ago.

“With the discovery of rock art dating to at least 67,800 years ago in Sulawesi, a large island on the most plausible colonization route to Australia, it is increasingly likely that the controversial date of 65,000 years for the initial peopling of Australia is correct,” Griffith University archaeologists Adam Brumm, a coauthor of the recent study, told Ars.

photo of an archaeologists studying a flashlight-lit cave wall adorned with ancient figures of animals in red — Archaeologists Shinatria Adhityatama studies a panel of ancient paintings in Liang Metanduno. Credit: Oktaviana et al. 2026

Archaeologists are still trying to work out exactly when, where, and how the first members of our species made the leap from the continent of Asia to the islands of Wallacea and, eventually, via several more open-water crossings, to Australia. Our picture of the process is pieced together from archaeological finds and models of ancient geography and sea levels.

“There’s been all sorts of work done on this (not by me), but often researchers consider the degree of intervisibility between islands, as well as other things like prevailing ocean currents and wind directions, changes in sea levels and how this affects the land area of islands and shorelines and so on,” Brumm said.

Most of those models suggest that people crossed the Makassar Strait from Borneo to Sulawesi, then island-hopped through what’s now Indonesia until they reached the western edge of New Guinea. At the time, lower sea levels would have left New Guinea, Australia, and New Zealand as one big land mass, so getting from New Guinea to what’s now Australia would actually have been the easy part.

A time capsule on the walls

There’s a sense of deep, deep time in Liang Metanduno. The cave wall is a palimpsest on which the ancient hand stencil is nearly covered by a brown-hued drawing of a chicken, which (based on its subject matter) must have been added sometime after 5,000 years ago, when a new wave of settlers brought domesticated chickens to the island. It seems almost newfangled against the ghostly faint outline of the Paleolithic hand.

A few centimeters away is another hand stencil, done in darker pigment and dating to around 21,500 years ago; it overlays a lighter stencil dating to around 60,900 years ago. Over tens of thousands of years, generations of people returned here with the same impulse. We have no way of knowing whether visitors 21,500 years ago, or 5,000 years ago might have seen a more vibrantly decorated cave wall than what’s preserved today—but we know that they decided to leave their mark on it.

And the people who visited the cave 21,500 years ago shared a sense of style with the artists who left their hands outlined on the wall nearly 40,000 years before them: both handprints have slightly pointed fingers, as if the artist either turned their fingertip or just touched-up the outline with some paint after making the stencil. It’s very similar to other hand stencils, dated to around 17,000 years ago, from elsewhere on Sulawesi, and it’s a style that seems unique to the island.

“We may conclude that this regionally unique variant of stencil art is much older than previously thought,” wrote Oktaviana and his colleagues.

photo of pointy-fingered hand stencils on a cave wall — These 17,000-year-old hand stencils from Liang Jarie Maros, in another area of Sulawesi, bear a striking resemblance to the much older ones in Liang Metanduno. Credit: OKtaviana et al. 2026

And Homo sapiens wasn’t the first hominin species to venture as far as Indonesia; at least 200,000 years earlier, Homo erectus made a similar journey, leaving behind fossils and stone tools to mark that they, too, were once here. On some of the smaller islands, isolated populations of Homo erectus started to evolve along their own paths, eventually leading to diminutive species like Homo floresiensis (the O.G. hobbits) on Flores and Homo luzonensis on Luzon. Homo floresiensis co-discoverer Richard Roberts has suggested that other isolated hominin species may have existed on other scattered islands.

Anthropologists haven’t found any fossil evidence of these species after 50,000 years ago, but if our species was in Indonesia by nearly 68,000 years ago, we would have been in time to meet our hominin cousins.

Nature, 2026. DOI: 10.1038/s41586-025-09968-y (About DOIs).

Kiona is a freelance science journalist and resident archaeology nerd at Ars Technica.

This 67,800-year-old hand stencil is the world’s oldest human-made art Read More »

Rocket Report: Chinese rockets fail twice in 12 hours; Rocket Lab reports setback

australia, Commercial space, Gilmour space, isar aerospace, launch, rocket report, Science, Space / Beth Washington / January 23, 2026

Another partially reusable Chinese rocket, the Long March 12B, is nearing its first test flight.

An Archimedes engine for Rocket Lab’s Neutron rocket is test-fired at Stennis Space Center, Mississippi. Credit: Rocket Lab

Welcome to Edition 8.26 of the Rocket Report! The past week has been one of advancements and setbacks in the rocket business. NASA rolled the massive rocket for the Artemis II mission to its launch pad in Florida, while Chinese launchers suffered back-to-back failures within a span of approximately 12 hours. Rocket Lab’s march toward a debut of its new Neutron launch vehicle in the coming months may have stalled after a failure during a key qualification test. We cover all this and more in this week’s Rocket Report.

As always, we welcome reader submissions. If you don’t want to miss an issue, please subscribe using the box below (the form will not appear on AMP-enabled versions of the site). Each report will include information on small-, medium-, and heavy-lift rockets, as well as a quick look ahead at the next three launches on the calendar.

Australia invests in sovereign launch. Six months after its first orbital rocket cleared the launch tower for just 14 seconds before crashing back to Earth, Gilmour Space Technologies has secured 217 million Australian dollars ($148 million) in funding that CEO Adam Gilmour says finally gives Australia a fighting chance in the global space race, the Sydney Morning Herald reports. The funding round, led by the federal government’s National Reconstruction Fund Corporation and superannuation giant Hostplus with $75 million each, makes the Queensland company Australia’s newest unicorn—a fast-growth start-up valued at more than $1 billion—and one of the country’s most heavily backed private technology ventures.

Homegrown rocket… “We’re a rocket company that has never had access to the capital that our American competitors have,” Gilmour told the newspaper. “This is the first raise where I’ve actually raised a decent amount of capital compared to the rest of the world.” The investment reflects growing concern about Australia’s reliance on foreign launch providers—predominantly Elon Musk’s SpaceX—to put government, defense, and commercial satellites into orbit. With US launch queues stretching beyond two years and geopolitical tensions reshaping access to space infrastructure, Canberra has identified sovereign launch capability as a strategic priority. Gilmour’s first Eris rocket lifted off from the Bowen Orbital Spaceport in North Queensland on July 30 last year. It achieved 14 seconds of flight before falling back to the ground, a result Gilmour framed as a partial success in an industry where first launches routinely fail.

The easiest way to keep up with Eric Berger’s and Stephen Clark’s reporting on all things space is to sign up for our newsletter. We’ll collect their stories and deliver them straight to your inbox.

Sign Me Up!

Isar Aerospace postpones test flight. Isar Aerospace scrubbed a potential January 21 launch of its Spectrum rocket to address a technical fault, Aviation Week & Space Technology reports. Hours before the launch window was set to open, the German company said that it was addressing “an issue with a pressurization valve.” A valve issue was one of the factors that caused a Spectrum to crash moments after liftoff on Isar’s first test flight last year. “The teams are currently assessing the next possible launch opportunities and a new target date will be announced shortly,” the company wrote in a post on its website. The Spectrum rocket, designed to haul cargoes of up to a metric ton (2,200 pounds) to low-Earth orbit, is awaiting liftoff from Andøya Spaceport in Norway.

Geopolitics at play... The second launch of Isar’s Spectrum rocket comes at a time when Europe’s space industry looks to secure the continent’s sovereignty in spaceflight. European satellites are no longer able to launch on Russian rockets, and the continent’s leaders don’t have much of an appetite to turn to US rockets amid strained trans-Atlantic relations. Europe’s satellite industry is looking for more competition for the Ariane 6 and Vega C rockets developed by ArianeGroup and Avio, and Isar Aerospace appears to be best positioned to become a new entrant in the European launch market. “I’m well aware that it would be really good for us Europeans to get this one right,” said Daniel Metzler, Isar’s co-founder and CEO.

A potential buyer for Orbex? UK-based rocket builder Orbex has signed a letter of intent to sell its business to European space logistics startup The Exploration Company, European Spaceflight reports. Orbex was founded in 2015 and is developing a small launch vehicle called Prime. The company also began work on a larger medium-lift launch vehicle called Proxima in December 2024. On Wednesday, Orbex published a brief press release stating that a letter of intent had been signed and that negotiations had begun. The company added that all details about the transaction remain confidential at this stage.

Time’s up... A statement from Orbex CEO Phil Chambers suggests that the company’s financial position factored into its decision to pursue a buyer. “Our Series D fundraising could have led us in many directions,” said Chambers. “We believe this opportunity plays to the strengths of both businesses, and we look forward to sharing more when the time is right.” The Exploration Company, headquartered near Munich, Germany, is developing a reusable space capsule to ferry cargo to low-Earth orbit and a high-thrust reusable rocket engine. It is one of the most well-financed space startups in Europe. Orbex is one of five launch startups in Europe selected by the European Space Agency last year to compete in the European Launcher Challenge and receive funding from ESA member states. But the UK company’s financial standing is in question. Orbex’s Danish subsidiary is filing for bankruptcy, and its main UK entity is overdue in filing its 2024 financial accounts. (submitted by EllPeaTea)

A bad day for Chinese rockets. China suffered a pair of launch failures January 16, seeing the loss of a classified Shijian satellite and the failed first launch of the Ceres-2 rocket, Space News reports. The first of the two failures involved the attempted launch of a Shijian military satellite aboard a Long March 3B rocket from the Xichang launch base in southwestern China. The Shijian 32 satellite was likely heading for a geostationary transfer orbit, but a failure of the Long March 3B’s third stage doomed the mission. The Long March 3B is one of China’s most-flown rockets, and this was the first failure of a Long March 3-series vehicle since 2020, ending a streak of 50 consecutive successful flights of the rocket.

And then… Less than 12 hours later, another Chinese rocket failed on its climb to orbit. This launch, using a Ceres-2 rocket, originated from the Jiuquan space center in northwestern China. It was the first flight of the Ceres-2, a larger variant of the light-class Ceres-1 rocket developed and operated by a Chinese commercial startup named Galactic Energy. Chinese officials did not disclose the payloads lost on the Ceres-2 rocket.

Neutron in neutral. Rocket Lab suffered a structural failure of the Neutron rocket’s Stage 1 tank during testing, setting back efforts to get to the inaugural flight for the partially reusable launcher, Aviation Week & Space Technology reports. The mishap occurred during a hydrostatic pressure trial, the company said Wednesday. “There was no significant damage to the test structure or facilities,” Rocket Lab added. Rocket Lab last year pushed the first Neutron mission from 2025 to 2026, citing the volume of testing ahead. The US-based company said it is now analyzing what transpired to determine the impact on Neutron launch plans. Rocket Lab said it would provide an update during its next quarterly financials, due in a few weeks.

Where to go from here?… The Neutron rocket is designed to catapult Rocket Lab into more direct competition with legacy rocket companies like SpaceX and United Launch Alliance. “The next Stage 1 tank is already in production, and Neutron’s development campaign continues,” the company said. Setbacks like this one are to be expected during the development of new rockets. Rocket Lab has publicized aggressive, or aspirational, launch schedules for the first Neutron rocket, so it’s likely the company will hang onto its projection of a debut launch in 2026, at least for now. (submitted by EllPeaTea)

Falcon 9 launches NRO spysats. SpaceX executed a late night Falcon 9 launch from Vandenberg Space Force Base on January 16, carrying an undisclosed number of intelligence-gathering satellites for the National Reconnaissance Office, Spaceflight Now reports. The mission, NROL-105, hauled a payload of satellites heading to low-Earth orbit, which are believed to be Starshield, a government variant of the Starlink satellites. “Today’s mission is the twelfth overall launch of the NRO’s proliferated architecture and first of approximately a dozen NRO launches scheduled throughout 2026 consisting of proliferated and national security missions,” the NRO said in a post-launch statement.

Mysteries abound… A public accounting of the agency’s proliferated constellation suggests it now numbers nearly 200 satellites with the ability to rapidly image locations around the world. The NRO has dozens more satellites serving other functions. “Having hundreds of NRO satellites on orbit is critical to supporting our nation and its partners,” the agency said in a statement. “This growing constellation enhances mission resilience and capability through reduced revisit times, improved persistent coverage, and accelerated processing and delivery of critical data.” What was unusual about the January 16 mission is it may have only carried two satellites, well short of the 20-plus Starshield satellites launched on most previous Falcon 9 launches, according to Jonathan McDowell, an astrophysicist and expert tracker of global space launch activity.

Long March 12B hot-fired at Jiuquan. China’s main space contractor performed a static fire test of a new reusable Long March rocket Friday, paving the way for a test flight, Space News reports. The test-firing of the Long March 12B rocket’s first stage engines occurred on a launch pad at the Dongfeng Commercial Space Innovation Test Zone at Jiuquan spaceport in northwestern China. The mere existence of the Long March 12B rocket was not publicly known until recently. The new rocket was developed by a subsidiary of the state-owned China Aerospace Science Technology Corporation, with the capacity to carry a payload of 20 metric tons to low-Earth orbit in expendable mode. It’s unknown if the first Long March 12B test flight will include a booster landing attempt.

Another one… The Long March 12B has a reusable first stage with landing legs, similar to the recovery architecture of SpaceX’s Falcon 9 rocket. The booster is designed to land downrange at a recovery zone in the Gobi Desert. The Long March 12B is the latest in a line of partially reusable Chinese rockets to reach the launch pad, following soon after the debut launches of the Long March 12A and Zhuque 3 rocket last month. Several more companies in China are working on their own reusable boosters. Of them all, the Long March 12B appears to be the closest to a clone of SpaceX’s Falcon 9. Like the Falcon 9, the Long March 12B will have nine kerosene-fueled first stage engines and a single kerosene-fueled upper stage engine. Chinese officials have not announced when the Long March 12B will launch.

Artemis II rolls to the launch pad. Preparations for the first human spaceflight to the Moon in more than 50 years took a big step forward last weekend with the rollout of the Artemis II rocket to its launch pad, Ars reports. The rocket reached a top speed of just 1 mph on the four-mile, 12-hour journey from the Vehicle Assembly Building to Launch Complex 39B at NASA’s Kennedy Space Center in Florida. At the end of its nearly 10-day tour through cislunar space, the Orion capsule on top of the rocket will exceed 25,000 mph as it plunges into the atmosphere to bring its four-person crew back to Earth.

Key test ahead… “This is the start of a very long journey,” said NASA Administrator Jared Isaacman. “We ended our last human exploration of the Moon on Apollo 17.” The Artemis II mission will set several notable human spaceflight records. Astronauts Reid Wiseman, Victor Glover, Christina Koch, and Jeremy Hansen will travel farther from Earth than any human in history as they travel beyond the far side of the Moon. They won’t land. That distinction will fall to the next mission in line in NASA’s Artemis program. This will be the first time astronauts have flown on the Space Launch System rocket and Orion spacecraft. The launch window opens February 6, but the exact date of Artemis II’s liftoff will be determined by the outcome of a critical fueling test of the SLS rocket scheduled for early February.

Blue Origin confirms rocket reuse plan. Blue Origin confirmed Thursday that the next launch of its New Glenn rocket will carry a large communications satellite into low-Earth orbit for AST SpaceMobile, Ars reports. The rocket will launch the next-generation Block 2 BlueBird satellite “no earlier than late February” from Launch Complex 36 at Cape Canaveral Space Force Station. However, the update from Blue Origin appears to have buried the real news toward the end: “The mission follows the successful NG-2 mission, which included the landing of the ‘Never Tell Me The Odds’ booster. The same booster is being refurbished to power NG-3,” the company said.

Impressive strides… The second New Glenn mission launched on November 13, just 10 weeks ago. If the company makes the late-February target for the next mission—and Ars was told last week to expect the launch to slip into March—it will represent a remarkably short turnaround for an orbital booster. By way of comparison, SpaceX did not attempt to refly the first Falcon 9 booster it landed in December 2015. Instead, initial tests revealed that the vehicle’s interior had been somewhat torn up. It was scrapped and inspected closely so that engineers could learn from the wear and tear.

Next three launches

Jan. 25: Falcon 9 | Starlink 17-20 | Vandenberg Space Force Base, California | 15: 17 UTC

Jan. 26: Falcon 9 | GPS III SV09 | Cape Canaveral Space Force Station, Florida | 04: 46 UTC

Jan. 26: Long March 7A | Unknown Payload | Wenchang Space Launch Site, China | 21: 00 UTC

Rocket Report: Chinese rockets fail twice in 12 hours; Rocket Lab reports setback Read More »

Tiny falcons are helping keep the food supply safe on cherry farms

birds, cherries, Ecology, food supply, kestrels, raptors, Science, syndication / Beth Washington / January 23, 2026

Campylobacter is a common cause of food poisoning and is on the rise in Michigan and around the world. It spreads to humans through food products made from, or that come into contact with, infected animals, primarily chickens and other birds. So far, only one outbreak of campylobacteriosis has been definitively linked to feces from wild birds. Still, because it causes milder symptoms than some other types of bacteria, the Centers for Disease Control considers campylobacter a significantly underreported cause of food-borne illness that may be more common than current data indicates.

“Trying to get more birds of prey would be beneficial to farmers,” Smith said. “If you have one predator, versus a bunch of prey, you have fewer birds overall. If you have a lot fewer birds, even if the ones that are there are carrying bacteria, then you can reduce the transmission risk.”

The study’s findings that kestrels significantly reduce physical damage and food safety risks on Michigan cherry farms demonstrate that managing crops and meeting conservation goals—by bolstering local kestrel populations and eliminating the need to clear wildlife habitat around agricultural areas—can be done in tandem, study authors say. They recommend farmers facing pest-management issues consider building kestrel boxes, which cost about $100 per box and require minimal maintenance.

Whether nesting boxes in a given region will be successfully inhabited by kestrels depends on whether there is an abundance of the birds there. In Michigan’s cherry-growing region, kestrels are so abundant that 80 percent to 100 percent of boxes become home for kestrels rather than other nesting birds, said Catherine Lindell, avian ecologist at Michigan State University and senior author of the study.

“It seems like this is just a great tool for farmers,” Lindell said, suggesting interested farmers “put up a couple boxes and see what happens.”

K.R. Callaway is a reporter and editor specializing in science, health, history, and policy stories. She is currently pursuing a master’s degree in journalism at New York University, where she is part of the Science, Health, and Environmental Reporting Program (SHERP). Her writing has appeared in Scientific American, Sky & Telescope, Fast Company and Audubon Magazine, among others.

This story originally appeared on Inside Climate News.

Tiny falcons are helping keep the food supply safe on cherry farms Read More »

Watch a robot swarm “bloom” like a garden

architecture, collective behavior, living architecture, robot swarms, robotics, Science, swarm intelligence / Shannon Garcia / January 21, 2026

Researchers at Princeton University have built a swarm of interconnected mini-robots that “bloom” like flowers in response to changing light levels in an office. According to their new paper published in the journal Science Robotics, such robotic swarms could one day be used as dynamic facades in architectural designs, enabling buildings to adapt to changing climate conditions as well as interact with humans in creative ways.

The authors drew inspiration from so-called “living architectures,” such as beehives. Fire ants provide a textbook example of this kind of collective behavior. A few ants spaced well apart behave like individual ants. But pack enough of them closely together, and they behave more like a single unit, exhibiting both solid and liquid properties. You can pour them from a teapot like ants, as Goldman’s lab demonstrated several years ago, or they can link together to build towers or floating rafts—a handy survival skill when, say, a hurricane floods Houston. They also excel at regulating their own traffic flow. You almost never see an ant traffic jam.

Naturally scientists are keen to mimic such systems. For instance, in 2018, Georgia Tech researchers built ant-like robots and programmed them to dig through 3D-printed magnetic plastic balls designed to simulate moist soil. Robot swarms capable of efficiently digging underground without jamming would be super beneficial for mining or disaster recovery efforts, where using human beings might not be feasible.

In 2019, scientists found that flocks of wild jackdaws will change their flying patterns depending on whether they are returning to roost or banding together to drive away predators. That work could one day lead to the development of autonomous robotic swarms capable of changing their interaction rules to perform different tasks in response to environmental cues.

The authors of this latest paper note that plants can optimize their shape to get enough sunlight or nutrients, thanks to individual cells that interact with each other via mechanical and other forms of signaling. By contrast, the architecture designed by human beings is largely static, composed of rigid fixed elements that hinder building occupants’ ability to adapt to daily, seasonal, or annual variations in climate conditions. There have only been a few examples of applying swarm intelligence algorithms inspired by plants, insects, and flocking birds to the design process to achieve more creative structural designs, or better energy optimization.

Watch a robot swarm “bloom” like a garden Read More »

Zillow removed climate risk scores. This climate expert is restoring them.

California, climate change, insurance risk, Science, syndication / Kelly Newman / January 21, 2026

In this way, climate risk models today are better suited to characterize the “ broad environment of risk,” said Chris Field, director of the Stanford Woods Institute for the Environment. “ The more detailed you get to be either in space or in time, the less precise your projections are.”

Matouka’s California climate risk plugin is designed for communicating what he said is the “standing potential risks in the area,” not specific property risk.

While climate risk models often differ in their results, achieving increased accuracy moving forward will be dependent on transparency, said Jesse Gourevitch, an economist at the Environmental Defense Fund. California is unique, since so much publicly available, state data is open to the public. Reproducing Matouka’s plugin for other states will likely be more difficult.

Private data companies present a specific challenge. They make money from their models and are reluctant to share their methods. “A lot of these private-sector models tend not to be very transparent and it can be difficult to understand what types of data or methodologies that they’re using,” said Gourevitch.

Matouka’s plugin includes publicly available data from the state of California and federal agencies, whose extensive methods are readily available online. Overall, experts tend to agree on the utility of both private and public data sources for climate risk data, even with needed improvements.

“People who are making decisions that involve risk benefit from exposure to as many credible estimates as possible, and exposure to independent credible estimates adds a lot of extra value,” Field said.

As for Matouka, his plugin is still undergoing beta testing. He said he welcomes feedback as he develops the tool and evaluates its readiness for widespread use. The beta version is available here.

Claire Barber is a fellow at Inside Climate News and masters in journalism student at Stanford University. She is an environmental and outdoor journalist, reporting primarily in the American Southwest and West. Her writing has appeared in The San Francisco Chronicle, Outside, Powder Magazine, Field & Stream, Trails Magazine, and more. She loves to get lost in the woods looking for a hot spring, backpacking to secluded campsites, and banana slugs.

This story originally appeared on Inside Climate News.

Zillow removed climate risk scores. This climate expert is restoring them. Read More »

Macaque facial gestures are more than just a reflex, study finds

macaques, Science / Mike M. / January 20, 2026

Based on the video analysis, scientists identified three facial gestures they wanted to focus on: the lipsmack macaques use to signal receptivity or submission; the threat face they make when they want to challenge or chase off an adversary; and chewing, a non-social, volitional movement. Then, using the fMRI scans, the team located key brain areas involved in triggering these gestures. And when this was done, Ianni and her colleagues went deeper—quite literally.

Under the hood

“We targeted these brain areas with sub-millimeter precision for implantation of micro-electrode arrays,” Ianni explains. This, for the first time, allowed her team to simultaneously record the activity from many neurons spaced across the areas where the brain generates facial gestures. The electrodes went into the primary motor cortex, the ventral premotor cortex, the primary somatosensory cortex, and the cingulate motor cortex. When they were in, the team once again exposed the macaques to the same set of social stimuli, looking for neural signatures of the three selected facial gestures. And that’s when things took a surprising turn.

The researchers expected to see a clear division of responsibilities, one where the cingulate cortex governs social signals, while the motor cortex is specialized in chewing. Instead, they found that every single region was involved in every type of gesture. Whether the macaques were threatening a rival or simply enjoying a snack, all four brain areas were firing in a coordinated symphony.

This led Ianni’s team to the question of how the brain distinguished between social gestures and chewing, since it apparently wasn’t about where the brain processed the information. The answer was in different neural codes—different ways that neurons represent and transmit information in the brain over time.

The hierarchy of timing

By analyzing neural population dynamics, the team identified a temporal hierarchy across the cortex in macaques. The cingulate cortex used a static neural code. “The static means the firing pattern of neurons is persistent across both multiple repetitions of the same facial gesture and across time,” Ianni explains, and maintained their firing pattern till 0.8 seconds after that. “A single decoder which learns this pattern could be used at any timepoint or during any trial to read out the facial expression,” Ianni says.

Macaque facial gestures are more than just a reflex, study finds Read More »