In an age when a single video can circle the globe in hours, a troubling new chapter in Japan’s battle with online falsehood has emerged, and it is unfolding not in some obscure corner of the internet but across the country’s most popular social platforms. According to Constella Security Japan Inc., a Tokyo-based cybersecurity firm, videos falsely portraying foreigners as stealing metal from railway tracks have been spreading widely across Japanese social media, apparently feeding a narrative that foreigners pose a threat to public safety. The firm’s investigation, released on a Saturday, revealed that these misleading clips, often accompanied by inflammatory Japanese captions such as “metal theft by foreigners” and “stealing metal from railway tracks,” had amassed roughly 4.3 million views across X, YouTube, and TikTok by September 15. The numbers alone are staggering, but what makes them even more unsettling is the method behind the deception: the videos were not recent footage of crimes in Japan at all. They were old clips filmed overseas, often in completely different contexts, stripped of their original meaning and repackaged with false narratives to fit a xenophobic agenda. In other words, ordinary people scrolling through their feeds were being shown a carefully edited version of reality, one designed to provoke fear, anger, and suspicion toward foreigners living in or visiting Japan.
The investigation into this digital sleight of hand began with a single viral post. In March, an X post appeared with the label “metal theft by foreigners,” showing people who appeared to be foreigners pulling cables away from railway tracks. The footage looked dramatic and damning, and it quickly gained traction. Yet Constella’s analysts soon discovered that the same footage had been posted on Instagram in Arabic as early as August of the previous year. It had circulated mainly in Arab countries and had absolutely nothing to do with Japanese railways, metal theft, or even Japan itself. When the company traced the video’s journey, it found that the original Arabic text had simply been removed and replaced with unrelated Japanese descriptions, a classic case of what researchers call “false context.” The same video also appeared on TikTok, often with similar misleading captions, and from there it seeped into the wider ecosystem of Japanese social media. By the time the clip had been viewed millions of times, the truth was lost in the noise. The original video, whatever its real purpose, had been weaponized. What makes this case particularly insidious is that it was not a single random post but part of a broader pattern: accounts that circulated the false information had, in many cases, previously posted messages opposing the acceptance of foreign workers. This was not merely misinformation; it was deliberate disinformation, manufactured to reinforce a pre-existing worldview.
What made the campaign even more disturbing, according to Constella’s analysis, was the nature of the accounts that helped spread the videos. The company examined accounts in Japan that had shared the false posts and found that many had a history of anti-foreign-worker sentiment. One account, with a substantial following of 70,000 users, had clearly established itself as a hub for such views long before the metal-theft video went viral. This suggests that the false video did not emerge from a vacuum; it was cultivated and amplified by a network of users who were already skeptical, if not openly hostile, toward immigration and foreign labor. In a society like Japan, where the foreign population has been steadily growing to address labor shortages, this kind of content could have real consequences. It feeds existing anxieties, paints an entire group of people with a single brush, and makes it harder for ordinary citizens to distinguish between isolated incidents and systemic problems. The fact that these accounts had such large audiences means the falsehood reached far beyond the usual echo chambers, seeping into the mainstream feeds of people who might not otherwise have encountered such views. And because the videos looked authentic—grainy, real, unpolished—they carried an emotional weight that a neutral article or official denial simply cannot match. The result is a social media environment where a videotaped fabrication can feel more convincing than a company’s investigation, and where fear travels faster than fact.
Constella’s analyst, Ryuna Nagamine, was blunt in her assessment of the tactics at play. She described the false-video campaign as using “a typical tactic of creating false information by adding explanations to unrelated videos to reinforce one’s own claims.” In other words, the creators of these posts understood that people are more likely to believe a claim if it is attached to something visual and immediate, even if the connection is entirely fabricated. The human brain is wired to trust what it can see, and video, especially raw-looking footage, triggers an instinctive “this is real” reaction. By attaching a false narrative to an unrelated clip, the manipulators bypassed the critical thinking that might otherwise kick in. Nagamine offered practical advice for viewers who come across such posts: consider the content of the video separately from the written text and verify whether it is factual. That may sound simple, but in practice it requires a kind of digital literacy that many people have not yet developed. It means pausing before sharing, asking where the video came from, and being willing to accept that a clip might be old, might be from another country, or might be entirely staged. The analyst’s warning is a reminder that the most dangerous pieces of misinformation are often the ones that feel the most real, because they are engineered to exploit our emotional responses rather than our rational minds.
This latest incident is far from an isolated anomaly in Japan’s recent history with disinformation. The country has seen similar falsehoods before, often in the wake of natural disasters, when fear and uncertainty are at their highest. After the massive earthquake and tsunami that devastated northeastern Japan in 2011, false claims about foreigners committing crimes spread rapidly, preying on a population already struggling to make sense of a catastrophe. The same pattern repeated after the earthquake that hit Ishikawa Prefecture in 2024, and more recently, weak-evidence antiforeign posts—such as exaggerated claims about the number of crimes committed by foreigners—have appeared during election periods. Each time, the specific claim is different, but the underlying structure is the same: a real social anxiety is identified, a false story is constructed to feed that anxiety, and social media does the rest. Nonprofit organizations and mainstream media outlets have repeatedly called on the public to be wary of unverified information, urging people to engage in fact-checking and to question what they see online. These calls, however, often struggle to keep pace with the speed and scale of digital dissemination. By the time a fact-check is published and spread, the false video may have already been viewed by millions, shared by thousands, and cemented into the minds of people who will never see the correction. This asymmetry—between the speed of lies and the slowness of truth—is one of the defining challenges of the modern information age, and Japan is not immune to it.
For ordinary viewers, the takeaway from this investigation is both sobering and hopeful. It is sobering because it shows how easily a foreigner’s face can be co-opted into a narrative of fear, how a piece of old footage from a distant country can be transformed into a weapon against a whole community. It is hopeful because it also shows that the truth can be uncovered, that careful analysis by cybersecurity firms and journalists can expose the machinery of deception, and that ordinary people, by adopting a more skeptical mindset, can resist manipulation. The human story behind the 4.3 million views is not really about railway tracks or metal theft at all. It is about how we, as individuals and societies, choose to relate to people who are different from us. When a false video confirms our existing biases, it feels comfortable, and we are tempted to share it. But when we pause, when we ask where a video came from and what it really shows, we not only protect ourselves from being deceived—we also protect our neighbors, our coworkers, and the strangers who live among us from being unjustly accused. In a country like Japan, which is slowly and sometimes reluctantly opening its doors to a foreign workforce, the stakes could not be higher. The next time a shocking video appears on social media, the most human response is not to react with fear, but to react with curiosity. Ask questions. Seek evidence. Remember that behind every clip is a context, and behind every caption is a choice. Choose the truth.