By now you’re probably overwhelmed enough by baby names to understand how parents could call their kid Apple or Blanket. They were tired and just pointed at something in the room. If you’re having trouble staying on top of naming trends from this century, don’t worry, this handy algorithm will at least help you come up with ideas for the next one.
Using 7,500 popular American baby names, developer and designer Nate Parrot trained a neural network to convert each name into numbers — a strategy used in machine learning. The algorithm then combined the numbers representing these existing names to create unfamiliar combinations. The result? Names that make kids seem like they crawled out of a Black Mirror episode.
Apparently, by analyzing features of common names, the computer could distinguish between words that seem more or less name-like. For instance, words like “happy” and “candle” were seen as more name-like than “automobile” and “larynx.” That’s fair. Even if it makes your kid sound like a damn hippie, it’s still more environmentally friendly than a car.
Sure, the following 30 baby names may fill you with the kind of future shock reserved for Herbie Hancock. Rest assured, though, you won’t be around to get weird about your great great great great grandkid being named Hhrsrrrrrr … unless they come up with an algorithm for that.
[H/T] Fast Company