Janelle Shane spent some time teaching a neural network how to generate recipes. She set it to learn from 30,000 existing recipes, but learning to cook is hard. After all, it can't taste the results. But even before the recipes are completed, it had a difficult time learning ingredients, measurements, and processes. The results are quite interesting. For example, here are some ingredients the machine suggests.
1 ½ teaspoon chicken brown water
1 teaspoon dry chopped leaves
1/3 cup shallows
10 oz brink custard
¼ cup bread liquid
2 cup chopped pureiped sauce
½ cup baconfroots
¼ teaspoon brown leaves
½ cup vanilla pish and sours
½ cup white pistry sweet craps
1 tablespoon mold water
¼ teaspoon paper
1 cup dried chicken grisser
15 cup dried bottom of peats
¼ teaspoon finely grated ruck
And this is a thing that it came up with repeatedly for some reason, and was quite adamant that I use:
1 cup plaster cheese
Shane also fed recipes into a different neural network that had already been trained on the works of H.P. Lovecraft.
Bake at 350 degrees for 30 to 32 minutes. Test corners to see if done, as center will seem like the next horror of Second House.
Whip ½ pint of heavy cream. Add 4 Tbsp. brandy or rum to possibly open things that will never be wholly reported.
Cook over a hot grill, or over glowing remains of tunnel mouth.
With blender on high speed, add ice cubes, one at a time, making certain each cube is the end.
Dice the pulp of the eggplant and put it in a bowl with the vast stark rocks.
NOTE: As this is a tart rather than a cheesecake, you should be disturbed.
She later fed her cooking network some text from H.P. Lovecraft to see what would happen. Yeah, that was just as funny. Read an archive of the experiments at Postcards from the Frontiers of Science. -via Metafilter
We hope you like this article!
Please help us grow by sharing: