First-hand information for everyone
In September 2026 the U.S. Department of War released a heavily redacted infrared (IR) video recorded from a military aircraft over the Middle East. The accompanying mission report described six small, roughly spherical objects moving at an estimated 480 miles per hour. A new analysis of the released footage finds a different, more prosaic pattern in the pixels: all six objects brighten and dim repeatedly at rates that match the wingbeat frequencies of flying animals.
The study measured the brightness of each object in every video frame and found steady periodic flicker at between 7.1 and 7.9 cycles per second (Hz). Each object beat at its own precise rate and out of phase with the others. To check whether the flicker could be an artifact of video compression or the thermal sensor, the authors ran control tests using planted, constant-brightness objects; those tests did not reproduce the observed flicker. That makes a processing or sensor glitch a less likely cause of the repeated brightening and dimming.
To put those flicker rates in context, the researchers applied the same analysis to ground-based IR recordings from the Galileo Project near Las Vegas. There, flying animals showed steady brightness modulations at 3.4–10.0 Hz — a range that overlaps well with the 7.1–7.9 Hz seen in the military video and with typical bird wingbeat frequencies. In other words, the temporal fingerprint of the six objects in the leaked footage is consistent with wingbeats.
Geometry of the encounter also matters. The paper shows that an object that is nearly stationary relative to the ground but sits at about 60–80% of the aircraft’s altitude would project onto the ground at an apparent speed near the reported 480 mph. In plain language, a small flock of birds flying closer to the aircraft could appear to cross the ground at very high speeds in a platform-mounted sensor, simply because the sensor and platform were moving.
Taken together, the authors conclude that the six objects are best explained as a small flock of birds flying nearer to the aircraft than the crew assumed — not as unknown high-speed craft. The analysis is careful to frame this as the most plausible explanation given the available, redacted video, not as an absolute proof. The paper notes that releasing additional platform and sensor metadata (for example: precise aircraft motion, sensor pointing, and timing information) would permit a definitive geometric and radiometric reconstruction and could either refute or strengthen this hypothesis.
The study also makes a methodological point for the wider public debate about unidentified aerial phenomena (UAP). Even when only heavily redacted imagery is available, reproducible, quantitative analysis of pixel-level behavior can test specific physical and biological hypotheses. At the same time, the authors emphasize limits: without the full, unredacted sensor and platform files, certain alternative explanations cannot be completely ruled out.
In short, this work does not claim the mystery is solved beyond doubt, but it presents a testable, terrestrial explanation that fits both the temporal flicker in the footage and what we know about bird wingbeat frequencies and observation geometry. The paper calls for the release of additional sensor and platform data where possible to allow independent verification and fuller resolution of such incidents.