Neural Networks Have Launched a New Wave of Scams on TikTok
A new "flower fever" is sweeping TikTok—and it's not about beauty, but a massive scam. Hundreds of videos featuring orchids with cat faces are deceiving users with claims like: "A magical variety," "Your neighbors don't have this," and "Just a couple of dollars for seeds." In reality, such plant varieties simply do not exist, and the videos are entirely generated by neural networks.
The funnel itself is incredibly straightforward: viral AI videos drive traffic to a pre-lander page where these "miracle seeds" are sold for $5–$15, but the delivered order turns out to be a package containing regular wheat or sawdust. The problem is that proving the scam is extremely difficult—technically, the seller did ship an order.
This tactic is far from new—back in 2024, this mass scam swept through the American audience. However, back then, the videos were crudely generated and low-quality. Now, with the advent of more powerful neural networks, the scam has been given a second life.
In December 2025, the media outlet Futurism published an article featuring dozens of examples of accounts flooding TikTok with these AI videos. Despite exposures from e-commerce experts and bloggers, the stream of deceived customers shows no signs of slowing down.
The CIS audience has also caught on to this trend. Dozens of videos featuring "miracle plants" in Russian and Ukrainian are popping up across Facebook and TikTok feeds. The target demographic for the scam is highly predictable: women aged 40+ who trust video content and struggle to distinguish AI generation from real footage. The purchase is entirely emotional and takes just a couple of minutes—see it, gasp, click, buy.
The flower scam isn't the first case of its kind in the physical products vertical. The packaging changes, but the mechanics remain identical: a beautiful visual, an impulse buy, and a low price point. Neural networks have simply accelerated this process, making it possible to churn out endless content in a matter of minutes. As a result, we can expect to see more of these setups across various verticals.
