What is the neuronet shown in the video supposed to do?

The neuronet is trained to recognize artificial objects in high-resolution lunar images. It was trained with Apollo landing-site hardware and then used to scan for Luna 9 and related components, flagging potential artifacts for further comparison with historical photos.

What did the neuronet actually identify in the data so far?

The system successfully detected the Luna 16 lander and the Surveyor 7 spacecraft in known locations, which validates the method. It then flagged a dark, geometrically shaped feature near the expected Luna 9 region and found related objects in additional images.

How certain are the findings that this is Luna 9?

The discovery is not claimed as definitive. The authors describe the match as about 70–80% and emphasize that they are reporting a possible identification, not a confirmed Luna 9 site.

What broader implications do these findings have about aliens or lunar artifacts?

The video argues that if alien observers ever visited the Solar System, the Moon would be an ideal place to observe, which motivates checking the Moon for anomalous artifacts. It also frames the discovery as a reminder of Cold War history and our ongoing space ambitions.

What does the video say about future Moon exploration by China or Europe?

The presenter notes China’s plans to build a Moon north-pole station by 2028 and suggests China may outpace the U.S. in this goal. It also jokes about European involvement and even cheese, highlighting a broader geopolitical context for Moon exploration.

Where can I learn more about science, AI, and circuits after watching this?

The video plugs Brilliant.org, describing it as a platform with interactive courses in science, computer science, and math. It offers a 30-day free trial and a discount via a channel link, inviting viewers to explore topics like digital circuits and AI.

What is the main takeaway from this Moon artifact discussion?

The main takeaway is that scientists can use modern imaging plus AI to search for historical artifacts on the Moon, potentially revisiting Cold War mysteries. Even if evidence remains uncertain, such work motivates ongoing exploration and international space ambitions.

The First Moon Landing Wasn’t Apollo — And We Just Found It

Sabine Hossenfelder

Science & Technology4 min read7 min video

Feb 28, 2026|323,361 views|15,601|1,755

hossenfelder science with sabine science humour science news sabine science news science humor moon ai neural net luna 9 moon probe space

Save to Pod

Key Moments

TL;DR

Luna 9 could predate Apollo; AI search hints at an artifact, but not confirmed.

Key Insights

Luna 9 (1966) reportedly landed before Apollo 11, but its exact landing site has never been confirmed.

NASA’s Lunar Reconnaissance Orbiter imagery shows Apollo hardware and other artifacts, yet Luna 9 hasn’t been positively identified in the data.

A neural network trained on known lunar hardware flagged a possible Luna 9 site, plus nearby debris, across multiple images, suggesting a real signal rather than an artifact.

The researchers stop short of proof, reporting a 70–80% match and calling for independent verification before declaring a discovery.

Beyond the science, the video uses the potential find to ponder moon exploration, aliens, and future space politics (Soviet early in the space race, then US vs China).

THE CLAIM: LUNA 9 BEFORE APOLLO AND ITS UNCERTAIN GROUND

The video begins by challenging the common narrative that Apollo 11 was the first successful moon landing. It asserts that Luna 9, a Soviet probe, actually achieved a successful landing in 1966—years before Apollo 11. However, the landing site has never been precisely identified, leaving room for doubt and conspiracy theories about whether the mission even landed. This setup introduces a Cold War-era mystery: a potential early lunar touchdown that modern science may be able to verify—or debunk—through new analysis.

FROM LRO IMAGES TO NEURAL NETWORK DETECTION

NASA’s Lunar Reconnaissance Orbiter has been surveying the moon since 2009, delivering imagery with resolutions around 25 cm per pixel. The speaker notes that this data reveals Apollo descent stages, rover tracks, and other hardware—yet Luna 9 remained elusive in those images. To tackle this, researchers trained a neural network on known lunar hardware locations (modules, rovers, and debris) to detect artificial objects. The system successfully identified Luna 16 lander and Surveyor 7 in prior tests, demonstrating the method’s viability before applying it to the Luna 9 region.

NEURAL NETWORKS AND ARTIFACT DETECTION: A POTENTIAL LUNA 9 SIGNAL

Applying the trained neuronet to the approximate Luna 9 landing area, the researchers detected a dark, oddly geometric feature in one image under favorable lighting. They then looked across eight other images and found the same object, which makes it unlikely to be imaging artifact. In addition, nearby images revealed several other features plausibly corresponding to ejected Luna 9 components. The team cross-checked these findings against old Luna 9 photos by reconstructing the horizon and landmarks to see if the new observations align, further supporting (but not proving) a potential match.

EVIDENCE IN CONTEXT: WHAT THE 70–80% MATCH MEANS

Despite the intriguing detections, the researchers emphasize that their match is not definitive. They report a 70–80% similarity to Luna 9 and deliberately avoid claiming a confirmed identification. The evidence is compelling enough to warrant careful follow-up, but the possibility that the feature could be another artifact or object remains real. This cautious stance reflects scientific rigor: exciting leads must be independently verified before making a confident claim.

ALIENS AND THE MOON: A THOUGHT EXPERIMENT

A notable part of the narrative is a speculative leap: if intelligent outsiders visited the solar system, they might have left artifacts on the Moon to observe humanity from a safe vantage. The video makes the case that scanning for strange technology on the Moon could yield extraordinary scientific and technological insights, even if the premise is unlikely. This line of thought frames the Moon as a potential archive of extraterrestrial visits and as a natural place to search for relics.

HISTORICAL REFLECTION AND GEOPOLITICAL FUTURES OF THE MOON

The discussion shifts to a broader historical and geopolitical arc. It notes that the Soviet Union was first on the Moon, a triumph increasingly forgotten as time passes and national nostalgia fades. The speaker then connects this history to current space ambitions: the United States faces a new space race with China, which plans to land on the Moon and build a station, potentially by 2028. The remark about Europe adds a light touch about optimism and regional goals, underscoring how the Moon remains a focal point of national prestige and scientific ambition.

LOOKING FORWARD: PERMANENT MOON PRESENCE AND EDUCATION

The video closes by tying the Moon’s exploration to practical paths forward, including the concept of a permanent lunar outpost to sustain long-term observation and discovery. It also pivots to an educational note, promoting Brilliant.org as a resource for building knowledge in science, math, and computer science. The sponsor message emphasizes bite-sized, adaptive learning that can scale with a learner’s background, and it offers a 30-day free trial plus a discount to viewers. This segment blends scientific curiosity with practical learning tools.

Mentioned in This Episode

●Tools & Products

Common Questions

The video explains that Luna 9 is claimed to have landed in 1966, years before Apollo 11. However, the exact landing site has never been confirmed, which is why researchers used a neuronet to search for possible evidence rather than asserting a definitive identification.