UFOs/UAPs: The Science

Science is a method, not a tribe. The method is: make observations, generate hypotheses, test them, publish the data, invite replication or refutation. The tribe is the social structure that decides which observations are worth making, which hypotheses are respectable, and which data gets published. On UAPs, the tribe and the method have been in conflict for decades. The method says: anomalous observations require investigation. The tribe says: this particular class of anomalous observations carries too much cultural baggage to investigate without embarrassment. What’s changed in the last five years is that a small number of credentialed scientists have decided the method matters more than the tribe.

Avi Loeb is the most visible. He is the Frank B. Baird Jr. Professor of Science at Harvard, former chair of the Harvard astronomy department, former chair of the National Academies board on physics and astronomy, and the author of over 900 peer-reviewed papers. He is not a person you can dismiss as credulous or marginally credentialed. In 2021 he founded the Galileo Project at Harvard, an instrumented observation program specifically designed to collect systematic data on UAPs and other anomalous aerial phenomena. The name is deliberate: Galileo was told by the institution of his day not to look through the telescope. Loeb thinks something similar is happening now, and he has the academic standing to say so and keep his job.

Instrumented Observation Changes the Game

The core problem with the historical UAP literature is that almost all the data is incidental. Pilots didn’t go up looking for UAPs; they encountered something unexpected and reported it after the fact. Radar operators weren’t running controlled experiments; they captured something anomalous in the course of ordinary operations. The resulting dataset is real but messy: chain-of-custody problems, instrument calibration uncertainty, human perception limitations, and selection bias toward the most dramatic incidents rather than the most representative ones.

What Loeb’s Galileo Project is doing is different in kind. He has deployed telescope arrays with multiple imaging systems, including radio, infrared, and optical sensors, along with audio detectors and weather instruments, at fixed observation sites. The goal is continuous, automated, calibrated data collection across multiple sensor modalities simultaneously. When something anomalous appears, you have corroborating data from independent instruments rather than a single sensor read or a human eyewitness. The first Galileo Project observing site went operational at Harvard in 2023.

Loeb’s team has already processed large volumes of data and identified objects that don’t match standard atmospheric or aircraft profiles. The published analyses are careful not to overclaim: they identify anomalies and describe the parameters of those anomalies (size, velocity, reflectivity) without asserting causes. This is methodologically correct. The job of first-pass instrumented observation is to build a characterized dataset, not to explain it. What the project has established is that the observation infrastructure can work, that anomalous data exists in the sky, and that systematic collection is possible. That’s the foundation everything else requires.

The SCU and What Rigorous Video Analysis Actually Shows

The Scientific Coalition for UAP Studies (SCU) is a smaller and less famous organization than the Galileo Project, but its work is worth understanding. The SCU is a non-profit composed of scientists, engineers, and former military and intelligence professionals who apply standard scientific methodology to UAP data. They publish peer-reviewed analyses in their own journal and collaborate with academic institutions.

Their 2019 analysis of the Aguadilla incident is one of the most technically detailed UAP analyses in the public record. The incident occurred in April 2013 over Puerto Rico. A Department of Homeland Security aircraft captured infrared footage of an object that appeared to enter and exit the ocean, split into two objects, and maintain consistent velocity through the water and air interface. The SCU analysis ran the video frame-by-frame, calculated the object’s velocity using the aircraft’s position data and the thermal camera’s field of view, and found the object was moving at approximately 40-120 mph over water and apparently transitioning between media (air and water) without deceleration.

The analysis is 161 pages. It accounts for camera motion, aircraft position, atmospheric conditions, and a range of conventional explanations. The conclusion is not “alien”; the conclusion is “this object’s behavior is anomalous by the physical parameters we measured, and no conventional explanation accounts for all the observed characteristics.” That’s a scientific conclusion. It’s specific, falsifiable, and grounded in the data. It can be wrong. But challenging it requires engaging with the methodology, not dismissing the subject category.

Metamaterials and the Chain-of-Custody Problem

One of the more contested claims in the UAP space involves alleged recovered materials with anomalous properties. Luis Elizondo, the former director of AATIP, has stated that materials were collected and analyzed as part of the program. Lue Cornet at BAASS (Bigelow Aerospace Advanced Space Studies) wrote a DIA technical report referencing “metamaterials” with properties outside known manufacturing capabilities. The Skinwalker Ranch connection (BAASS operated there, the DIA contracted with them) adds a layer of murk that makes sober analysis harder.

What can be said with confidence: at least one peer-reviewed paper exists in the materials science literature describing the analysis of an object alleged to have anomalous origin (a bismuth-magnesium-zinc layered structure with properties consistent with a high-frequency waveguide), published by a scientist named Garry Nolan at Stanford, who has published over 300 peer-reviewed papers and is a legitimately serious immunologist and data scientist. The paper doesn’t claim alien origin. It describes the material’s properties, notes they’re unusual, and invites further analysis. This is how science is supposed to work.

The chain-of-custody problem is real and significant. Without verified provenance, a piece of exotic material is interesting but not conclusive. Materials with anomalous properties exist in terrestrial manufacturing; the question is whether this particular material’s specific combination of properties and layering is consistent with known processes. Nolan’s conclusion is that it’s unusual enough to warrant more study. That’s a modest claim, and it’s the right one given the evidence.

What Physics Actually Requires for the Anomalies to Be Real

It’s worth doing the physics explicitly, because the dismissal often rests on an unstated assumption that whatever these objects are, they’d have to violate known physics to be real. That assumption is wrong in several specific ways.

The Nimitz object’s reported behavior, dropping from 80,000 feet to sea level in under a second, implies acceleration somewhere around 40,000g. That’s incompatible with human-piloted craft because human physiology fails above roughly 10g for sustained periods. It’s not incompatible with an unmanned vehicle with a propulsion mechanism we haven’t publicly identified, particularly if the craft uses field propulsion rather than reaction mass. The Alcubierre drive, a legitimate solution to Einstein’s field equations proposed by Miguel Alcubierre in 1994, describes a geometry of spacetime that would allow a craft to move faster than light relative to distant observers without violating local physics, because the craft isn’t moving through spacetime; spacetime is moving around it. The drive requires exotic matter with negative energy density, which we haven’t demonstrated manufacturing at scale, but the physics is valid. More practically: several aerospace engineers have proposed non-reaction-mass propulsion concepts involving electromagnetic field interactions with spacetime that would produce the observed acceleration profiles without exotic matter. None of these are operational technologies in the public domain. None are physically impossible.

The trans-medium behavior observed in the Aguadilla incident (an object apparently transitioning from air to water without deceleration) is similarly anomalous but not physically forbidden if the object is generating a supercavitation effect or uses a field that interacts differently with each medium. Supercavitation is real: the Russian Shkval torpedo uses a gas envelope to reduce water resistance and achieves speeds over 200 mph underwater. The Aguadilla object’s behavior requires a more sophisticated version of this principle, but sophisticated is not impossible.

The point is that applying physics rigorously to the observed anomalies doesn’t rule them out; it identifies what class of technology or phenomenon would produce them, and that class is “advanced but not impossible” rather than “physically forbidden.” That’s a meaningful constraint that the science can work with. You can ask: what technology produces these characteristics? What energy requirements does it imply? What signatures would it leave that we could detect with better instrumentation? Those are answerable scientific questions. The Galileo Project is building the infrastructure to answer them.

Limited Data Is Not No Data

The methodological mistake in the mainstream dismissal is treating “limited data” as equivalent to “no data.” They’re not the same, and conflating them is a failure of scientific reasoning. When you have an anomalous observation with a robust chain of custody, multiple independent sensor confirmations, analysis by credentialed researchers published in peer-reviewed venues, and a characterized parameter space that rules out known explanations, you have something. It’s limited. It doesn’t prove a specific cause. But it’s real data about a real phenomenon, and the appropriate response is continued investigation with better instrumentation, not dismissal.

The history of science is full of anomalous observations that were dismissed by the institutional mainstream and turned out to be pointing at something real. Ball lightning was considered folklore for decades; it’s now an accepted physical phenomenon. Prions were dismissed as impossible (proteins that self-replicate without nucleic acid? absurd) until the evidence became overwhelming. The institutional dismissal of anomalous observation has a bad track record. That’s not an argument for credulity; it’s an argument for actually looking, which is what Loeb and the SCU are doing.

What’s different about UAPs from ball lightning or prions is the cultural baggage, the decades of hoaxes, pop culture contamination, and bad-faith claims that have made the subject radioactive in polite academic company. That baggage is real and it has real costs: it makes it harder to get funding, harder to get papers through peer review, harder to attract researchers who want a conventional career. But the baggage is a sociological problem, not a scientific one. It doesn’t touch the question of whether anomalous aerial phenomena with anomalous physical characteristics exist. On that question, the data that exists says yes.

The Verdict: Gold

The science is thin. The Galileo Project is new. The SCU dataset is limited. The metamaterial chain-of-custody is imperfect. None of this is the kind of robust, multi-decade, heavily replicated literature that we associate with established science. But the absence of robust literature is itself partly the product of the institutional taboo, not just the absence of a real phenomenon. When you don’t fund investigation, you don’t get data. When you don’t get data, you can point to the absence of data as justification for not funding investigation. That’s circular, and it’s been the loop that’s kept UAP science marginal for seventy years.

What the Galileo Project and the SCU represent is the beginning of a break from that loop: credentialed scientists, using legitimate methods, applying actual rigor to anomalous data, publishing results, and inviting scrutiny. That’s the correct process. It should have started decades earlier. It’s starting now. The data so far is anomalous enough to justify the investigation and not conclusive enough to settle what’s being investigated. That’s exactly where you’d expect to be at the beginning of a genuine scientific inquiry into a real phenomenon. It’s also where you’d be if the phenomenon turns out to be explainable and the anomalies resolve under better instrumentation. Either way, you need the data. Now someone is collecting it.

Loeb has said something in interviews worth taking seriously: that the reason scientists avoid this subject isn’t that they’ve evaluated the evidence and found it wanting; it’s that they haven’t evaluated the evidence at all, because doing so would cost them professionally. That’s an epistemological scandal, not a scientific conclusion. Science without observation is just peer-reviewed opinion. The Galileo Project is the attempt to replace opinion with data, conducted by someone who understood exactly what it would cost him in credibility to do it and did it anyway. Whatever the data eventually shows, that’s how science is supposed to work.