Was interstellar comet 3I/ATLAS an alien probe?

What was observed?

On July 1, 2025, the NASA-funded ATLAS survey telescope in Río Hurtado, Chile, first reported it to the Minor Planet Center. The discovery team measured an orbital eccentricity of about 6.1, a strongly open path. Its orbit is tilted about 175 degrees, so it lies close to the plane of the planets but runs the opposite way. It reached its closest point to the Sun at the end of October 2025, just inside the orbit of Mars.

It was active from the start. The discovery team’s images showed a compact coma, the fuzzy cloud of gas and dust around a comet’s icy core. Hubble saw a teardrop-shaped cocoon of dust and a plume from the Sun-warmed side, much like our own comets. On August 6, 2025, the Webb telescope found carbon dioxide, water, carbon monoxide and water ice, plus a tentative sign of carbonyl sulfide. Its ratio of carbon dioxide to water was among the highest ever seen in a comet. Spectra from the Very Large Telescope in Chile showed nickel gas but, at first, no iron.

More than a dozen NASA missions observed it, along with ESA’s Mars orbiters and its Jupiter-bound Juice spacecraft. After it passed the Sun, Webb detected methane, the first direct detection of that gas on an interstellar visitor, NASA says, and measured about 30 times more deuterium, a heavy form of hydrogen, than in solar system comets. Its path also bent very slightly away from the course gravity alone predicts, a small push called non-gravitational acceleration.

What did people infer from it?

NASA and ESA drew the plain inference: a comet from another star. NASA’s FAQ says its characteristics, color, speed and direction all fit a comet, and ESA says its shape and behavior indicate one. At a NASA briefing on November 19, 2025, associate administrator Amit Kshatriya said the agency is always looking for life beyond Earth, “but 3I/Atlas is a comet,” PBS News reported.

A second inference got more attention. On July 16, 2025, Adam Hibberd, Adam Crowl and Avi Loeb posted a paper titled “Is the Interstellar Object 3I/ATLAS Alien Technology?” They pointed to its path lying within about 5 degrees of the planets’ plane (a chance they put near 0.2%), to passes near Venus, Mars and Jupiter (about 0.005% or less), and to its being hidden behind the Sun, seen from Earth, near its closest point to the Sun, when a spacecraft could brake unseen. They also counted a lack of comet gas in its spectrum and a size too large for an asteroid. They raised the idea that it could be technological and even hostile. Yet they called the paper “largely a pedagogical exercise,” and its conclusion says a completely natural object, probably a comet, is by far the most likely outcome, while adding that the work offered “compelling insights” into the technology idea.

Loeb, a Harvard professor who heads the Galileo Project, kept a public list of anomalies on his blog. By late November 2025 it included a forecast pass by Jupiter at almost exactly Jupiter’s Hill radius, roughly the farthest distance at which Jupiter’s gravity can hold a moon, which he said might mean it intends to release devices there; more nickel than iron in its gas, which he compared to industrial nickel alloys; a jet pointing toward the Sun; a very massive nucleus; extreme negative polarization of its light (a pattern in how its light waves line up); and very fast brightening near the Sun. He rated it 4 on his 0-to-10 Loeb Scale, which runs from a natural comet or asteroid (0) to alien technology that could threaten humanity (10). Levels 8 to 10 are kept for confirmed artificial objects, so a 4 marks it as anomalous, not artificial. He also wrote, more than once, that it is most likely a natural object, but that a black swan, a small chance with huge stakes, deserves a close look. In April 2026, he still rated it 4, while noting that its fading light showed no sign of artificial light.

Does the evidence support the alien-probe idea?

So far, no. Each argument has been checked against data.

• Comet gas showed up. Within weeks, Webb found a gas cloud dominated by carbon dioxide, and Hubble put its solid core at no more than 5.6 kilometers across. That answered the July paper’s point about missing comet gas and undercut its estimate that, if it were an asteroid, it would be about 20 kilometers wide.
• Nickel gas is not new for comets. In 2021, the European Southern Observatory reported iron and nickel gas, in roughly equal amounts, in about 20 solar system comets, even far from the Sun, and nickel gas around 2I/Borisov. For 3I/ATLAS, a 2026 study, not yet peer reviewed, found strong nickel and iron emission after it passed the Sun, with the ratio of nickel to iron moving toward values seen in solar system comets.
• The sunward jet has a natural model. Loeb himself co-wrote a peer-reviewed paper with Eric Keto explaining the early sunward glow as icy grains that last longer on the sunlit side, where stronger evaporation throws out bigger grains.
• The extra push is small and fits outgassing. NASA says the small changes in its path fit what escaping gas does to comets. By our arithmetic, the push in JPL’s orbit fit is roughly a thousand times smaller than the one the July paper said an intercept of Jupiter would need.
• The odds were figured after the fact. Jason Wright, director of the PSETI Center at Penn State, argued on his blog that the probabilities were computed for features picked after they were already known. An analogy: any poker hand you are dealt had a tiny chance of coming up, but that alone does not show the deck was stacked. The analogy has a limit: stacked decks exist, so a pattern predicted in advance would count. Later coincidences, such as the Jupiter distance, face the same question: predicted in advance, or noticed afterward?
• No artificial radio signals turned up. The SETI Institute’s Allen Telescope Array listened from 1 to 9 GHz and found no signals worth following up. Breakthrough Listen used the Green Bank Telescope on December 18, 2025, and found no transmitter broadcasting in all directions stronger than about 0.1 watt, a tenth of a cell phone’s output. The FAST telescope listened on four dates and found no credible signal. MeerKAT in South Africa found no transmitter either, but it did detect radio lines from hydroxyl (OH), a molecule of one oxygen and one hydrogen atom, and they behaved as its observers expected for a comet. Silence does not rule out a quiet probe, but narrowband radio is how our own Voyager probes talk.

Isn’t it closed-minded to rule out aliens?

This is the strongest objection. We have seen only three interstellar objects, so nobody knows what normal looks like for them. Researchers who search for signs of alien technology take them seriously: a 2025 review by James Davenport and colleagues, posted on arXiv, argues that checking them for technosignatures must become routine as new telescopes find more.

But keeping a question open is different from treating an oddity as evidence. The same review urges care before calling something a technosignature mainly because it differs from solar system objects. An object from another star should differ, and Wright notes that no two comets seem quite alike. The checks were done, and they found a comet.

What do people often get wrong? “Anomalous” does not mean “artificial.” It means “not yet explained.” And the data are not locked away. NASA’s open-data page points to public archives for Webb, TESS, SPHEREx and MAVEN data. Officials said a federal government shutdown had kept them from answering every theory, PBS reported. For the wider question of secrecy, see our answer on whether governments are hiding evidence of aliens. If a video claims to show a spaceship, our guide on checking viral videos can help.

What is still unresolved?

Quite a lot. Nobody knows exactly how big the solid core is: Hubble limits as of August 2025 put it between 440 meters and 5.6 kilometers across. The Webb team suggested its ices saw more radiation than those of our comets, or formed near the zone in its star’s disk where carbon dioxide freezes. In November 2025, Wright listed its high nickel, extreme polarization, odd water level and fast brightening as puzzles that interest planetary scientists.

Its exact time and place of origin remain highly uncertain, the Webb team wrote in 2026. But Webb found little carbon-13, a heavier form of carbon that builds up in galaxies over time, and with the high deuterium that points to a very old, cold birthplace. In a Nature paper published in June 2026, the team says it may have formed as long as 12 billion years ago, while noting significant uncertainties in that age. Our Sun formed 4.5 billion years ago. ESA said its teams were analyzing Juice’s observations, and in April 2026 Loeb was still waiting for data from NASA’s Juno spacecraft at Jupiter.

How can you check this yourself?

• Read the paper’s own conclusion. The July 2025 paper is free on arXiv; read its final section.
• Look up the orbit. Search “3I” in NASA JPL’s Small-Body Database to see its open orbit and the measured non-gravitational push.
• Separate the four steps. For any new claim, ask what was observed, what was inferred, whether the evidence supports it, and what is still open. Our guide to extraordinary claims walks through this.
• Put it in context. See what official reviews found about UFO reports, and read our answer on the Fermi paradox.

Asking whether 3I/ATLAS was a probe was fair. Every test so far says it was a comet from another star, and a strange and fascinating one.