Computers may be replacing a whole lot of human jobs, but we don't think they're quite ready for writing songs for the holidays. A research scientist trained her neural network on 240 different Christmas carols to see what the artificial intelligence system could learn about the song Rudolph the Red-Nosed Reindeer. She asked it what comes after the line "Rudolph the Red-Nosed Reindeer" and it responded with several renditions, including one that started off like this:
Rudolph the Red Nosed Reindeer, with its red belly
The All-gracious King of all the earth,
Had a baby at Christmas time, On Christmas Day,
A true and holy Deity, Went down to earth,
With human flesh for sacrifice...
Research scientist Janelle Shane also asked her neural network to produce an original carol. The AI seemed to focus on the commercialism of the holiday, writing a song about buying a yacht "now."
"I can't tell whether it has misunderstood what Christmas carols are all about, or understood some of them only too well," she wrote in a blog post Tuesday about her experiment.
Shane, who last Christmas neural networks, which are modeled on the human brain and are able to determine individual rules and responses after being fed and trained on large amounts of data.-- it came up with "Merry Hunga Poppers," "Grandma's Spritches" and "Apricot Dream Moles" -- has also . The researcher has been using self-learning
She previously used them to generate superhero names ("Superbore" and "Red Fart"), new phrases for Valentine's Day candy hearts ("sweat poo," love 2000 hogs yeah" and "stank love") and new Dungeons and Dragons creatures and spells (like "Summon Ass" and "Shield of Farts").