3I/ATLAS: Comet or Alien Probe? What the Data Show

Two weeks later, a Harvard professor asks whether it was built. Fourteen months, more than a dozen spacecraft and one appointment at Jupiter after that, it is possible to say quite precisely where the mystery lies. What this article gives you: the full timeline, what the telescopes measured, what Avi Loeb actually wrote, how his 22 "anomalies" held up, and where 3I/ATLAS is right now. What it does not give you: a verdict built on headlines. The mystery is real, but it sits somewhere other than most people expected.

The next day there is more. Observers in Chile and the United States point their telescopes at the dot, others dig through older images and find it as early as June 14. The positions yield an orbit, and the orbit is wrong, at least for anything that belongs to our solar system. On July 2 at 21:31 Universal Time, the Minor Planet Center in Cambridge, Massachusetts, issues circular MPEC 2025-N12. It states it plainly: a strongly eccentric, hyperbolic orbit, eccentricity around 6. Plus first, cautious reports of a faint haze around the nucleus and a tail just three arcseconds long.

Among the observers credited with those reports is one who has been here before: R. Weryk. Robert Weryk discovered ʻOumuamua in Hawaii in October 2017, the first object proven to have come to us from another star. Now the third one has arrived. It is named 3I/ATLAS: "3" for the third, "I" for interstellar. And before the world knows what it is made of, a Harvard professor will ask whether it was built.

This is the story of the fourteen months that followed. It has two narrators: the instruments of more than a dozen spacecraft and large telescopes, and an astrophysicist who argued with them. By the end you can say fairly exactly where the mystery lies. It is not where most of the headlines suggested.

An orbit that never comes back

What makes an object interstellar? Not its appearance, but its speed. Everything firmly bound to our Sun travels on an ellipse: planets, asteroids, even the comets from the distant Oort Cloud that need hundreds of thousands of years for a single lap. The eccentricity of an ellipse is below 1. For 3I/ATLAS, NASA's Jet Propulsion Laboratory today gives a value of 6.14. That is no longer an ellipse, and not even a sharply bent hyperbola. It is nearly a straight line: the Sun only nudges the visitor slightly off course.

Far out, before the Sun sped it up, it was traveling at about 58 kilometers per second, roughly 36 miles every second. That is nearly twice as fast as the second interstellar visitor, comet 2I/Borisov in 2019. NASA translates it into everyday units: about 137,000 miles per hour (221,000 km/h) at discovery, and about 153,000 mph (246,000 km/h) at its closest point to the Sun. According to the European Space Agency, it came from the direction of the disk of the Milky Way, and it will never return.

Speed like that says something about age. Over billions of years, stars pick up velocity because they keep passing gas clouds and other stars and get nudged each time. A team led by Matthew Hopkins at Oxford placed 3I/ATLAS into a model of the Milky Way as early as July 2025. The result: its speed points to an age of more than 7.6 billion years. Our Sun is 4.6 billion years old. If the model is right, this chunk of ice and dust formed before Earth existed. Tom Statler, NASA's lead scientist for small solar system bodies, said in November 2025 that nobody can know for sure, but the thought gave him goosebumps: 3I/ATLAS is not only a window into another solar system, but into a time before the Earth and the Sun formed.

It is worth keeping track of which parts are measurement and which are model. The orbit and the speed are measured. The age is an estimate derived from how old stars moving at that speed typically are. It is well reasoned, but it is a probability, not a birth certificate.

Comets have always unsettled people because they arrive unannounced. For thousands of years they were read as signs: of a ruler's death, of war, of upheaval. On the Bayeux Tapestry, which tells the story of the Norman conquest of England in 1066, men point up at a long-tailed star; we now know it was Halley's Comet. The modern era turned the omen into a math problem. A handful of positions gives you an orbit, and the orbit tells you where a body came from and where it is going. For 3I/ATLAS, that calculation showed within a day that it came from another star. What it actually is, the math could not say. That took the months that followed, and a predecessor that had already stumped science once.

The predecessor: ʻOumuamua and the light sail

On October 19, 2017, the Pan-STARRS1 telescope on Haleakalā on Maui picks up a faint dot moving too fast. It gets the Hawaiian name ʻOumuamua, roughly "a messenger from afar arriving first." It is already on its way out when it is found, and it is strange: no haze, no tail, but strong swings in brightness that point to a very elongated or very flat shape.

In June 2018, a team led by Marco Micheli publishes a detailed analysis of the orbit in Nature. On its way out of the solar system, ʻOumuamua was pushed a tiny bit harder than gravity alone explains. That kind of extra acceleration is familiar from comets: when ice vaporizes, the gas acts like a weak thruster. Only, with ʻOumuamua, no gas could be seen.

Into that gap, in fall 2018, stepped a paper by Shmuel Bialy and Abraham "Avi" Loeb, then chair of Harvard's astronomy department. They calculated whether the pressure of sunlight could explain the acceleration. It could, if ʻOumuamua were an extremely thin sheet, less than a millimeter thick. The paper ends with a thought that went around the world: ʻOumuamua might be debris from technology, or, in a "more exotic scenario," a fully operational probe sent intentionally into the vicinity of Earth by an alien civilization.

The field pushed back almost unanimously, and it did so the way that counts in science: with counter-models. The most convincing so far appeared in March 2023, also in Nature. Jennifer Bergner and Darryl Seligman showed that hydrogen, which builds up over long periods as cosmic rays strike a comet's ice, can escape invisibly when the ice warms and produce exactly this small push. No haze, because hydrogen gas does not glow. That is not proven either; ʻOumuamua has long been out of reach.

Astronomy took one lesson from the episode, and Loeb took another. Astronomy's lesson: when the next visitor comes, find it early, so there is time to measure it. Loeb's lesson: do not rule out the technological possibility before the data do. With 3I/ATLAS both lessons collided, and this time there were eight months of observing time to work with.

July 16, 2025: "a pedagogical exercise"

Two weeks after the discovery, on July 16, 2025, a paper appears on the preprint server arXiv under the title "Is the Interstellar Object 3I/ATLAS Alien Technology?" The authors are Adam Hibberd and Adam Crowl of the Initiative for Interstellar Studies in London, and Avi Loeb. The first sentence of the abstract sets out how the paper is meant, and it should be read literally.

"As largely a pedagogical exercise, in this paper we present additional analysis into the astrodynamics of 3I/ATLAS, and hypothesize that this object could be technological, and possibly hostile as would be expected from the 'Dark Forest' resolution to the 'Fermi Paradox'."

Adam Hibberd, Adam Crowl, Abraham Loeb: "Is the Interstellar Object 3I/ATLAS Alien Technology?", abstract, arXiv:2507.12213, July 16, 2025.

The "Dark Forest" idea comes from science fiction, from a novel by the Chinese author Liu Cixin, the second book of the trilogy that opens with The Three-Body Problem. It holds that civilizations hide in the cosmos because any other could be a threat, which is why we hear nothing from them. A scientific paper that explicitly calls itself an exercise and borrows a motif from a novel: that is the starting point of the spaceship theory, and it is worth knowing before you read the headlines that grew out of it.

The paper lists oddities in the orbit. 3I/ATLAS moves retrograde, against the direction the planets orbit, and yet almost in their plane, tilted by barely five degrees. It passes comparatively close to Venus, Mars and Jupiter, which the authors consider very unlikely. And at its closest point to the Sun it sits, as seen from Earth, behind the Sun, where it cannot be observed. From this they build not a claim but scenarios, and the scenarios are phrased so they can be tested. An optimal intercept course toward Earth would mean arrival in late November or early December 2025. A particular extra acceleration would suggest an intent to steer toward Jupiter and rendezvous with it after perihelion.

On the same day, Loeb explained on his blog on Medium how he himself rated the odds: by far the most likely outcome is that 3I/ATLAS is a completely natural interstellar object, probably a comet. The hypothesis, he wrote, was an interesting exercise in its own right and fun to pursue regardless of how likely it was to be true. That matters for any fair assessment. In his own writing, Loeb did not claim that 3I/ATLAS was a spaceship. He kept a possibility open and attached numbers to it. The headlines, "spaceship," "mothership," "hostile probe," were made out of that possibility, not out of a measurement.

The reaction from his colleagues was sharp all the same. Samantha Lawler, an astronomer at the University of Regina in Canada, told Live Science that the vast majority of researchers follow the principle that extraordinary claims require extraordinary evidence, and that the evidence presented was "absolutely not extraordinary." The main methodological objection concerned the probabilities. If you first know the properties of an object and then calculate how unlikely exactly that combination is, you will find something improbable about any object. Astronomer Jason Wright of Penn State, who himself runs a center for the search for extraterrestrial intelligence, summed up the objection in November 2025: the arithmetic is right, but the features were chosen after they were already known, a textbook error in reasoning about probability.

Both sides deserve some credit. The critics are right that after-the-fact probabilities prove very little. Loeb is right that predictions can settle an argument. And the predictions were now on the table, with dates attached.

What the telescopes saw: carbon dioxide, nickel, a haze

On August 6, 2025, the James Webb Space Telescope turns toward 3I/ATLAS, which at that point is still 3.3 astronomical units from the Sun, more than three times as far out as Earth. The team led by Martin Cordiner of NASA's Goddard Space Flight Center finds a cloud of gas and dust around the nucleus, a coma, made up mostly of carbon dioxide. Plus water, carbon monoxide, carbonyl sulfide and water ice. The ratio of carbon dioxide to water is 7.6, one of the highest ever measured in a comet. The researchers interpret it cautiously: the ice may have absorbed more radiation than the ice of our comets, or the body formed close to the zone where carbon dioxide froze out in its home system.

At almost the same time, a team working with the European Southern Observatory's Very Large Telescope in Chile reports something stranger. The light of the coma shows lines of nickel, but none of iron. In our comets the two metals usually appear together. The researchers, led by Rohan Rahatgaonkar, also watch the amount of nickel climb steeply the closer the comet gets to the Sun. Their explanation is unspectacular: the nickel is not boiled off metallic chunks but released with little energy from dust grains, possibly via volatile nickel compounds. The second interstellar comet, Borisov, had also given off nickel.

The Hubble Space Telescope delivers an upper limit on size. It cannot see the nucleus through the haze, but the brightness caps how large it can be. NASA currently gives a diameter between 1,440 feet and 3.5 miles (440 meters to 5.6 kilometers). Later calculations based on the orbit, for example by John Forbes and a colleague in December 2025, land at about half a mile to two-thirds of a mile, roughly one kilometer.

By late summer 2025, the key question was settled, at least for most specialists: 3I/ATLAS has a coma, it has a tail, and it gives off the substances comets give off. It does so in unusual proportions, which is what you would expect from a body born around another star. Loeb saw it differently. He pointed out that the ratio of nickel to iron resembled industrially produced nickel alloys. Online, that remark quickly became a "metal spaceship." The measurement data say nothing of the sort. They only say that one spectral line is missing.

Want to know what is real and what is dyed glass in the moldavite market? Our guide Moldavite: Real or Fake, and What It Actually Is walks through it.

The list: 22 anomalies, ranked by likelihood

On November 23, 2025, Loeb publishes a ranked list on his blog titled "Anomalies of 3I/ATLAS, Organized by Likelihood." It is split into three classes, and each anomaly carries a probability. At the very top, with P = 0.00004: the predicted distance from Jupiter on March 16, 2026, 53.445 million kilometers plus or minus 0.06, is identical to Jupiter's Hill radius, the boundary inside which Jupiter's gravity dominates over the Sun's. Below that: the finely tuned arrival time near Mars, Venus and Jupiter; the large mass compared with ʻOumuamua and Borisov; a jet pointing toward the Sun, a so-called anti-tail; the nickel without iron; the orbit lying almost in the plane of the planets. In the second class: 3I/ATLAS came within nine degrees of the direction from which the famous "Wow!" signal was received in 1977 by Ohio State University's Big Ear radio telescope. By March 2026, the list has grown to 22 items.

The list is an unusual document because it does something you rarely see: it weighs things. It makes every claim individually open to attack. That is exactly what happened to it. Jason Wright went through Loeb's earlier list of ten point by point and reached a conclusion that should be stated fairly: four of the items genuinely interested comet researchers. The large amount of nickel, the extreme polarization of reflected light, the low water fraction and the rapid brightening. All four, he wrote, are the kind of deviation you would expect from a new kind of comet. He quotes a line that circulates among planetary scientists: comets are like cats. They have tails, and they do exactly what they want.

On the "Wow!" signal, Wright is blunt: nine degrees off means 3I/ATLAS did not come from that direction. For scale, nine degrees on the sky is about the width of eighteen full Moons side by side. And on the nickel, he points out that there is no fixed iron-to-nickel ratio in comets; the values vary widely. What remains are real oddities that can be described as comet physics, and probabilities that were only calculated once the properties were known.

October to December 2025: Mars, the Sun, Earth

On October 3, 2025, 3I/ATLAS passes Mars at just under 18 million miles (29 million km). Spacecraft orbit the red planet, and they get a close look. ESA's ExoMars Trace Gas Orbiter photographs it with its CaSSIS camera, which was actually built for the bright Martian surface a few hundred miles below. "The comet is approximately 10,000 to 100,000 times fainter than our usual target," said Nick Thomas, the camera's principal investigator. The images show a slightly fuzzy white dot: the nucleus inside its coma, a few thousand miles across. NASA's Mars Reconnaissance Orbiter captured it on October 2 with its HiRISE camera.

On October 29, 3I/ATLAS reaches the point of its orbit closest to the Sun, perihelion, 1.36 astronomical units out, between the orbits of Earth and Mars. From Earth it sits close behind the Sun during those weeks, just as Hibberd, Crowl and Loeb had flagged as an anomaly. It is not unobserved, though. ESA's Jupiter probe Juice, on its way to its target, switches on five instruments in November. On November 2, four days after perihelion, its spectrometer measures 2,000 kilograms of water vapor per second, about 4,400 pounds, or roughly 70 Olympic swimming pools a day. Most of it is released on the sunward side, much of it from icy dust grains in the haze. Gas and dust stretch at least three million miles. In April 2026, ESA sums up the results in a single phrase: "extreme but not exotic."

On November 19, 2025, NASA holds a press conference with images from many of its own spacecraft and telescopes. Amit Kshatriya, NASA associate administrator and one of the agency's most senior officials, addresses "the rumors" head-on. "This object is a comet," he says. "It looks and behaves like a comet, and all evidence points to it being a comet." But it comes from outside the solar system, he adds, and that makes it fascinating and scientifically very important.

On December 19, 3I/ATLAS makes its closest approach to Earth: about 168 million miles (270 million km), 1.8 astronomical units, nearly twice as far away as the Sun. The "optimal intercept" of Earth that the July paper had calculated for late November or early December does not happen. There was never any danger, as NASA's own information page also states.

Worth watching: NASA's full press conference from November 19, 2025, with the spacecraft images and the agency's statements in its own words.

March 16, 2026: the appointment at Jupiter

That left the entire weight of the technology hypothesis on a single date. Loeb's highest-ranked anomaly was the distance from Jupiter, and the July paper had floated a possible intent to rendezvous with Jupiter after perihelion. Four days before the date, on March 12, 2026, Loeb writes a kind of interim verdict under the title "3I/ATLAS is Fading Away." If nothing unusual happened near Jupiter, he wrote, he would regard 3I/ATLAS, which had shown comet activity, as only slightly less anomalous than ʻOumuamua. On his own scale for interstellar objects, running from 0 for natural to 10, he had given it a 4 at discovery.

Worth watching: Avi Loeb talking with Danny Jones in late 2025 about why he considered the Jupiter flyby the decisive test. What actually happened on March 16 follows right below.

On March 16, 2026, at 12:22 Universal Time, 3I/ATLAS passes Jupiter at a distance of 53.6 million kilometers, about 33.3 million miles. That is what the Jet Propulsion Laboratory's orbital data show today, calculated from thousands of position measurements. The distance really is close to Jupiter's Hill radius; the figure Loeb had quoted in November from an earlier orbit solution was about 100,000 miles (160,000 km) too small, and so fell outside the uncertainty he himself had given. More important is what happened next: nothing, as far as anything has been published. No swinging into orbit, no braking, no change of course beyond the small extra acceleration that was already known. The Juice images from the fall show a comet with two tails that, according to ESA, behaved like a typical comet of our own solar system despite its origin.

The extra acceleration itself is real, but small, and it has found natural explanations. Florian Neukart calculated in late 2025 that ordinary outgassing of carbon monoxide can account for it if the nucleus has a radius of a third of a mile to about two miles (half a kilometer to three kilometers). Forbes and his colleague worked from the mass the comet must have lost and arrived at a diameter of about one kilometer, in line with the other size estimates. Jupiter did not hold on to 3I/ATLAS. In September 2026 it is more than twice as far from the Sun as Jupiter.

Water from a colder world

The biggest mystery 3I/ATLAS left behind was not on anyone's list in 2025. Six days after perihelion, as the comet was just coming out from behind the Sun, the ALMA radio telescope array in Chile's Atacama Desert observed it. Unlike most optical instruments, radio telescopes can point toward the Sun. A team led by doctoral student Luis Salazar Manzano and astronomer Teresa Paneque-Carreño of the University of Michigan was looking for a rare form of water: semi-heavy water, in which one of the two hydrogen atoms is replaced by deuterium, hydrogen with an extra neutron in its nucleus.

In the comets of our solar system, there is roughly one of these semi-heavy molecules for every ten thousand ordinary water molecules. In 3I/ATLAS, the deuterium share is at least thirty times higher than in our comets and more than forty times higher than in Earth's oceans. It is the first measurement of its kind on an interstellar object, published in April 2026 in the journal Nature Astronomy. Deuterium builds up in ice when it is very cold. The researchers conclude that 3I/ATLAS must have formed at below roughly 30 kelvin, about −406 °F (−243 °C), and under weaker radiation than our comets. "This is proof that the conditions under which our solar system formed are not universal," Paneque-Carreño said. "That may sound obvious, but it is one of the things you have to prove."

Loeb had already commented on the first deuterium values, measured with the Webb telescope, on March 24, 2026. He noted that deuterium is used as fuel for nuclear fusion and asked whether the abundance could be a technological signature. The obvious natural explanation, he wrote, was ruled out. Assigning 3I/ATLAS to the rare group of old, metal-poor stars was "untenable," because such stars would not supply enough heavy elements. And ancient planet-forming disks could not have been colder than the cosmic background radiation of their time.

On September 11, 2026, a little over two weeks ago, the answer appeared on arXiv; it has since been accepted by The Astrophysical Journal Letters. Kenji Furuya and seven colleagues, among them Martin Cordiner, whose team had also measured the semi-heavy water with Webb, simulate the life of water ice from the start of a gas cloud to the dense core in which a star forms. Their result: it is precisely an environment poor in heavy elements, with at most half the amount found in our cosmic neighborhood today, that most easily produces this high deuterium share. Less carbon monoxide means that an intermediate product of deuterium enrichment survives longer. And the ratio of heavy methane to heavy water measured in 3I/ATLAS sits close to the value found in comet 67P/Churyumov–Gerasimenko, the comet visited by the Rosetta spacecraft.

That is the most remarkable twist in the whole story. The scenario Loeb declared untenable in March is, six months later, the one that best reproduces the measurement. It is not proof; it is a model that fits. But it tells a single coherent story together with the unusually low share of heavy carbon-13 that Webb measured: a comet from an old, cold, metal-poor corner of the Milky Way.

Where is 3I/ATLAS now?

Search engines reveal what a lot of people want to know: "is 3I/ATLAS still here," "is 3I/ATLAS coming back," "where is it now." The answer can be recalculated every day. On September 28, 2026, according to the Jet Propulsion Laboratory's orbital data, 3I/ATLAS is 11.7 astronomical units from the Sun, about 1.09 billion miles (1.75 billion km), and moving away from it at almost 59 kilometers per second, about 37 miles per second. In late July 2026 it crossed the distance of Saturn's orbit, the planet whose strange polar storm we cover in Saturn's Decagon: What Hubble Found at the South Pole. Its brightness has fallen to magnitude 23, within reach only of the largest telescopes in the world. It is not coming back. In roughly a century and a half it will be as far out as the inner edge of the Oort Cloud, and then, as before it arrived, it will drift between the stars.

For science, the episode is not over; it is just beginning. Data from eight months of observation will be analyzed for years. The Juice teams have announced that more measurements are still to be worked through. And the new Vera C. Rubin Observatory in Chile, which released its first images in 2025, has also delivered images of 3I/ATLAS. With Rubin, the next visitor will likely be found earlier than any of the three so far.

Seen from the ground, comets have always carried meaning beyond their chemistry. If you are more drawn to what people have read into lights in the sky, from falling stars to wishes, our piece on shooting star meaning and the meteor showers to watch covers the folklore side.

What is verified, and what is not

• Verified: 3I/ATLAS comes from another star. Its orbit is hyperbolic with an eccentricity of 6.14, it was traveling at about 36 miles per second far from the Sun, and it is leaving the solar system for good. It never came closer to Earth than about 168 million miles.
• Verified: It is active like a comet, with a coma, tails, carbon dioxide, water, carbon monoxide and nickel, in unusual proportions. Its water contains at least thirty times as much deuterium as that of our comets.
• Verified: Avi Loeb raised the possibility of alien technology, attached probabilities to 22 anomalies and named the Jupiter flyby as a test. What is verified is that he said these things, not their content. By all published data, the flyby was unremarkable.
• Traditional: Since antiquity, comets have been read as omens of upheaval; the Bayeux Tapestry shows Halley's Comet of 1066. That is cultural history, not a statement about 3I/ATLAS.
• Not proven: That 3I/ATLAS is a probe, a spaceship or artificial at all. No measurement has turned up a clue that cannot be described as comet physics.
• Not proven: Its exact age and its home. More than seven billion years and a cold, metal-poor birthplace are the models that fit best. Neither has been measured.

The open question

Anyone who followed 3I/ATLAS as a spaceship story may come away disappointed. The Jupiter date came and went, Earth was not targeted, the nickel comes from dust, the heavy water from cold. But the mystery has not disappeared. It has moved.

It no longer lies in the question of who sent this object, but in where the world it formed in came from. Somewhere in the Milky Way, possibly more than seven billion years ago, there was a cloud of gas and dust, colder and poorer in heavy elements than the one our Sun came from. Around a star we will never find, ice gathered, and a chunk of it was flung out long before Earth existed. It spent half the history of the universe between the stars and brushed past our solar system for a few months. In those months, we weighed its water.

Which star that was, whether it still shines, whether planets orbit it, whether anyone there ever looked up at the sky: 3I/ATLAS does not answer. It only says, with measured numbers, that our solar system is not the rule but one case among many. Loeb was right that unsolved mysteries deserve to be acknowledged. It is just that the mystery that remains is not a technological one. It is a cosmic one, and it is older than the Earth.

If you want another sky mystery that has been measured instead of merely talked about, read Hessdalen Lights: What 40 Years of Data Show, the Norwegian valley where unexplained lights have been under instrument watch since the 1980s.

Frequently asked questions

Is 3I/ATLAS an alien spaceship?

By everything that has been measured, no. 3I/ATLAS shows a coma and tails and gives off carbon dioxide, water, carbon monoxide and nickel, like a comet. NASA, ESA and the large majority of specialists call it an interstellar comet. Harvard astrophysicist Avi Loeb raised the possibility of alien technology and wrote himself that a natural origin was by far the most likely outcome. There is no evidence of an artificial origin.

Where is 3I/ATLAS now, and will it come back?

On September 28, 2026, 3I/ATLAS is about 11.7 astronomical units, roughly 1.09 billion miles, from the Sun according to NASA's Jet Propulsion Laboratory, farther out than Saturn's orbit. It is on an open, hyperbolic path, moving outward at almost 37 miles per second, and it will not return. Only the largest telescopes can still see it.

How big is 3I/ATLAS?

No telescope has seen the nucleus itself, because it is hidden in its cloud of gas. Hubble images put the diameter between about 1,440 feet and 3.5 miles (440 meters to 5.6 kilometers). Calculations from its small extra acceleration and its mass loss point to about one kilometer, a little over half a mile.

What is special about the water in 3I/ATLAS?

The ALMA radio telescope found at least thirty times as much deuterium, heavy hydrogen, in the water of 3I/ATLAS as in the comets of our solar system, and more than forty times as much as in Earth's oceans. That suggests it formed at below about 30 kelvin. A model published in September 2026 shows that an environment poor in heavy elements explains this value best.

Was 3I/ATLAS ever a danger to Earth?

No. 3I/ATLAS made its closest approach to Earth on December 19, 2025, at about 168 million miles (270 million km), nearly twice as far away as the Sun. NASA stated from the start that there was no danger.

What are Avi Loeb's 3I/ATLAS anomalies?

Since July 2025, Avi Loeb collected unusual features of 3I/ATLAS, ranked them by likelihood on November 23, 2025, and expanded the list to 22 items by March 2026, including the orbit's alignment, nickel without iron, a sunward jet and the distance from Jupiter. Other astronomers consider most items after-the-fact probabilities or properties you would expect from a comet born around another star. What is verified is that Loeb made these statements, not that they point to technology.

Sources and further reading

• Minor Planet Center: MPEC 2025-N12: 3I/ATLAS = C/2025 N1 (ATLAS), July 2, 2025. Discovery on July 1, four 30-second exposures, eccentricity about 6, observers Ye, Deen, Weryk.
• NASA Science: 3I/ATLAS Facts and FAQs. Speeds, perihelion, Earth distance, size from Hubble data, no danger.
• NASA Science: Comet 3I/ATLAS. Overview page with images from the participating missions.
• JPL Solar System Dynamics: Horizons. NASA's orbit calculator; distances for the table and the status on September 28, 2026 (object C/2025 N1).
• M. J. Hopkins et al.: From a Different Star: 3I/ATLAS in the context of the Ōtautahi-Oxford interstellar object population model, ApJ Letters 990, 2025. Age estimate above 7.6 billion years.
• M. Micheli et al.: Non-gravitational acceleration in the trajectory of 1I/2017 U1 (ʻOumuamua), Nature 559, 2018.
• S. Bialy, A. Loeb: Could Solar Radiation Pressure Explain ʻOumuamua's Peculiar Acceleration?, ApJ Letters, November 2018. Light sail, "fully operational probe."
• J. B. Bergner, D. Z. Seligman: Acceleration of 1I/ʻOumuamua from radiolytically produced H2 in H2O ice, Nature, March 2023.
• A. Hibberd, A. Crowl, A. Loeb: Is the Interstellar Object 3I/ATLAS Alien Technology?, arXiv, July 16, 2025. Quote in the box above; Earth and Jupiter scenarios.
• H. Baker, Live Science, via Space.com: Controversial paper questions whether interstellar visitor 3I/ATLAS is 'possibly hostile' alien tech, July 2025. Statements by Loeb (blog) and Samantha Lawler.
• M. A. Cordiner et al.: JWST detection of a carbon dioxide dominated gas coma surrounding interstellar object 3I/ATLAS, ApJ Letters 991, 2025. CO2/H2O = 7.6.
• R. Rahatgaonkar et al.: Very Large Telescope Observations of Interstellar Comet 3I/ATLAS II, ApJ Letters, 2025. Nickel without iron, cyanide.
• J. T. Wright, AstroWright (Penn State): Loeb's 3I/ATLAS "Anomalies" Explained, November 9, 2025. Loeb's list of ten quoted in full with commentary.
• ESA: ESA's ExoMars and Mars Express observe Comet 3I/ATLAS, October 7, 2025. Quote from Nick Thomas.
• ESA: Five things Juice has revealed about Comet 3I/ATLAS, April 2, 2026. "Extreme but not exotic," 2,000 kg of water vapor per second.
• P. Pester, Live Science: '3I/ATLAS is a comet': NASA finally releases new 3I/ATLAS images, November 19, 2025. Quotes from Amit Kshatriya and Tom Statler.
• F. Neukart: Non-Gravitational Acceleration in 3I/ATLAS, arXiv, November/December 2025. Carbon monoxide outgassing as the explanation.
• J. C. Forbes et al.: The size of 3I/ATLAS from non-gravitational acceleration, arXiv, December 2025. Diameter about 820 to 1,050 meters.
• NRAO: 3I/ATLAS Contains 30X More Semi-Heavy Water Than Comets In Our Solar System, April 23, 2026. ALMA measurement, Salazar Manzano, Paneque-Carreño (Nature Astronomy).
• University of Michigan via ScienceDaily: Interstellar comet 3I/ATLAS contains strange water never seen in our solar system, May 8, 2026. Quotes from Paneque-Carreño.
• K. Furuya et al.: High water D/H ratio of the interstellar object 3I/ATLAS is consistent with a low-metallicity origin, arXiv, September 11, 2026, accepted by ApJ Letters.
• Avi Loeb, blog on Medium (wording checked, no link): "Is the Interstellar Object 3I/ATLAS Alien Technology?", July 16, 2025 · "Anomalies of 3I/ATLAS, Organized by Likelihood", November 23, 2025 · "3I/ATLAS is Fading Away, Leaving Us to Ponder Over Its 22 Mysterious Anomalies", March 12, 2026 · "The Anomalously High Abundance of Deuterium in 3I/ATLAS", March 24, 2026.
• Worth watching, but not cited as evidence: Avi Loeb in conversation with Danny Jones (Danny Jones Clips, late 2025) and NASA's press conference of November 19, 2025, both embedded above.

This article reports the state of published measurements as of September 28, 2026. Where it describes hypotheses, it labels them as such.

The visitor was a comet. The question it leaves behind is bigger than a spaceship.