When disaster strikes, the first thing people do is reach for their phones. They want to know if their loved ones are safe, where the flooding is worst, and how to get help or offer it. But in the chaos after the devastating floods along the Nepal–Tibet border, many people scrolling through social media found something else: a confusing, frightening swirl of fake videos, manipulated images, and fraudulent donation appeals. A BBC News report dated 31 August 2026 revealed that Nepalese authorities had had enough. They ordered Meta and TikTok to remove online content they considered misinformation about the floods, and they did so with an urgency that reflected just how dangerous the situation had become. It is one thing to see a misleading post during ordinary times; it is quite another when lives are hanging in the balance. The government’s move was not about censorship or controlling the narrative. It was about trying to stop the spread of lies that could cost people their lives, their money, or their last shred of hope.
The scale of the problem became clear in the days following the floods. As rescue teams worked to reach stranded communities and aid workers struggled to assess the damage, social media platforms were flooded with a different kind of debris: manipulated material, outright falsehoods, and content generated by artificial intelligence that looked disturbingly real. Some of this was simply cruel, like AI-synthesised videos that falsely claimed to show real-time disaster footage, often depicting scenes that had never happened or were from entirely different events. Other content was more sinister. Fraudulent donation QR codes began circulating, designed to divert emergency relief funds away from legitimate charities and into the pockets of scammers. These codes looked official, and in the panic and generosity that follow any major catastrophe, people scanned them without thinking. The Nepali government realised that this was not just an annoyance. It was a direct threat to the relief operation itself. Every dollar stolen was a dollar that would not reach a family in need. Every piece of fake footage was a reason for people to distrust real information, making the job of emergency responders even harder.
In response, Nepalese authorities formed a multi-agency task force led by the Nepal Telecommunications Authority, with representatives from the Ministry of Communication and Information Technology joining the effort. Their mission was straightforward but difficult: coordinate with the platforms and get the harmful content taken down as quickly as possible. Officials held discussions with representatives from Meta and TikTok, explaining that the usual process—where problematic posts are reported, reviewed, and then removed within 24 to 48 hours—was simply not good enough in the middle of a humanitarian emergency. During a flood, a day can make the difference between life and death. A false report that a dam has broken can send thousands of people fleeing into danger. A fake video of a collapsed bridge can cause panic and gridlock, blocking roads needed by ambulances and supply trucks. The authorities asked platforms to take proactive action, not just wait for complaints, and to detect and remove harmful content before it ever appeared on people’s screens. To their credit, the platforms did not refuse. Instead, they provided designated points of contact so that Nepalese officials could flag problematic content directly, bypassing the slow, impersonal reporting systems that ordinary users have to deal with. It was a small but significant step, acknowledging that emergency situations require special procedures.
At the same time, the Ministry of Information and Communications issued an official press release aimed not at the platforms, but at the public. The message was simple and deeply human: please stop sharing things you are not sure about. The Ministry urged people to refrain from circulating graphic photographs, unverified reports, or AI-generated rumours that could incite panic or worsen the trauma already being experienced by victims’ families. There was a particular sensitivity here. In the age of social media, we have become accustomed to seeing everything, but that does not mean we should share everything. A blurry photo of an injured person might be someone’s mother, father, or child. A rumour about a village being wiped out might be completely false, but the fear it causes is real. The Ministry emphasised that all news about damage, casualties, and relief efforts should rely strictly on authorised governmental sources. This may sound bureaucratic, but in a crisis, it is a lifeline. Official sources are not perfect, but they are accountable. They have a responsibility to be accurate. The same cannot be said for a random post from an unknown account, especially one that might have been created by a bot or an AI programme designed to spread confusion.
Why does all this matter? Because false information during a disaster is not a harmless nuisance. It causes tangible, real-world harm. It incites panic, and panic kills. It diverts critical resources away from the people who need them most, as emergency teams are forced to chase down false leads or respond to situations that never existed. It exploits the generosity of ordinary people, turning their desire to help into an opportunity for theft. And it deepens the trauma of survivors, who may see horrifying images of their own communities and not know whether what they are looking at is real. The situation in Nepal is not unique. It is a glimpse into a future that every country will face, where environmental disasters and synthetic media intersect. Climate change is making floods, fires, and storms more frequent and more severe. At the same time, artificial intelligence is making it easier than ever to create convincing fake images and videos. Put those two trends together, and you have a perfect storm. The question is whether social media platforms are operationally and technically equipped to respond. The 24-to-48-hour turnaround time that works for ordinary misinformation is a luxury that disaster responders simply do not have. If platforms cannot move faster, they will not just be failing their users; they will be failing the basic test of humanity.
This episode should serve as a wake-up call, not just for Nepal, but for the entire world. Governments need to think about how they will handle misinformation before the next disaster, not after. Platforms need to build systems that can detect synthetic media and scams in real time, especially during emergencies. And ordinary people need to become more careful digital citizens. It is natural to want to share information when you see something alarming or inspiring. But it is also necessary to pause, check the source, and ask whether the thing you are about to share is true and helpful. The Nepali government’s decision to confront Meta and TikTok directly was a brave and necessary move. It showed that public safety must come before corporate convenience. It also showed that the fight against misinformation is not just about algorithms and content moderation policies. It is about protecting real people, in real time, from harm. As the waters recede and the rebuilding begins, the lessons of this disaster should not be forgotten. The next flood, storm, or earthquake is already on its way. The question is whether we will be ready to face it without being drowned in lies.