First the robots came to dispense your cannabis, and now they seem to be enjoying some of the supply themselves. Amid rising concerns that artificial intelligence and automation is killing jobs, humans can at least rest easy knowing they’ll still be needed in a post-SKYNET world to come up with paint colors that don’t sound like a bizarre mix of juvenile Mad Libs humor and stoner references.
Janelle Shane, a research scientist, trained a neural network to concoct new paint colors by feeding it a list of nearly 8,000 Sherwin-Williams hues and their core paint values. The resulting data was…interesting, to say the least. According to Ars Technica:
“The longer it processed the dataset, the closer the algorithm got to making legit color names, though they were still mostly surreal: ‘Soreer Gray’ is a kind of greenish color, and ‘Sane Green’ is a purplish blue. When Shane cranked up ‘creativity’ on the algorithm’s output, it gave her a violet color called ‘Dondarf’ and a Kelly green called ‘Bylfgoam Glosd.’ After churning through several more iterations of this process, Shane was able to get the algorithm to recognize some basic colors like red and gray, ‘though not reliably,’ because she also gets a sky blue called ‘Gray Pubic’ and a dark green called ‘Stoomy Brown.'”
But before you run out to your nearest hardware store and request that they whip up a gallon of paint so you can slather your kitchen walls with “Gray Pubic,” check out the other colors this AI burped out:
At least half of these colors sound like they could double as strains. I don’t know about you, but if I walked into my neighborhood dispensary and saw jars labeled “Snowbonk,” “Grass Bat,” “Stanky Bean,” “Stoner Blue,” or the simple-yet-neutral “Dope,” I’d ask for an eighth and be on my merry way.
Shane concluded that her neural network “really likes brown, beige, and grey,” but “has really, really bad ideas for paint names.” That may be true, but considering some of the most popular cannabis varieties are Chocolope, Jillybean, and Blue Magoo, it has really, really good ideas for strain names.