June 19, 2018
The future of meming is now.
Most science these days is done to push harmful agendas like global warming or invent new niggertech to entertain the bugman masses. But once in a while, somehow, these four-eyes come up with something that truly advances the human condition.
While the most valuable commodity in human society used to be oil, this has long been replaced with something even more fundamental to human welfare: memes.
As such, the automation of meme production is going to trigger a fourth industrial revolution, which will propel our race into a glorious new future.
Artificial intelligence can do some remarkable things, from driving cars to improving breast cancer diagnosis. And now it can make memes.
Okay, so it’s shitty memes. But you gotta start somewhere. Probably better than my first memes – but don’t tell anybody.
Abel Peirson and Meltem Tolunay, two researchers at Stanford University, recently posted a paper to the arXiv preprint server detailing a new machine learning model they created that’s capable of creating pretty convincing memes:
All that processing power, and they could only make them 300p. Sad.
Peirson and Tolunay trained their machine learning algorithm with a dataset of more than 400 types of memes with multiple captions that they pulled from memegenerator.com using a Python script. To simplify the task, they focused only on “advice animal” style memes, the kind where an image of a specific character like “socially awkward penguin” is overlaid with a caption of text that represents the traits of that character, usually with a humorous observation.
They had to use normie memes to keep things simple. Alt-Right memes are way too big-brained for a basic AI system to handle.
After a few hardware revisions, the AI might be able to come up with good virgin v. chad memes.
“This allows for relatively simple collection of datasets,” the researchers wrote. “In this paper, we specifically refer to meme generation as the task of generating a humorous caption in a manner that is relevant to the initially provided image, which can be a meme template or otherwise.”
Yeah, that’s the basic bitch meme technology. Next, you’ll need to get the AI to combine multiple images together.
And finally, the ability to generate completely new memes from scratch using advanced artistic abilities.
There’s still a long way to go. But the future is looking bright indeed.
After training the algorithm on what kinds of captions go with which images, they were able to test it out by having it generate memes over and over. Some of them didn’t make a lot of sense:
Those pixels… Awful.
But others were pretty solid:
The impressive results will start when they start plugging this system straight into /pol/ and begin producing some really dank stuff.
I can only assume that the goal of this research is to provide tools for automated trolling, which is a noble cause. Right now, the biggest bottleneck in the trolling industry is definitely the labor requirements – giving troll engineers the tools to automate their work is sure to cause an explosive growth period in the sector, which is sure to boost the American economy like nothing we’ve ever seen.
Especially exciting is the prospect of implementing this system here at the Daily Stormer. The creation of memes is currently by far the biggest burden on our budget.
It’s estimated that the original Pepe created for this article cost us over 0.732 bitcoin. Donate now – memes don’t come cheap.
Some people might speculate that the left might utilize algorithmic meme generators to finally get their hands on quality material, turning the tide of the internet war in their favor.
Fortunately, this is impossible.
Making funny commie memes is mathematically impossible, and would definitely break any computer that attempts it – like a division by zero.
“Tech support? This is Aaron Goldberg. I tried to get my computer to crank out some anti-Nazi memes, and now smoke is coming out of it. Oy vey!” “Yeah, sir, you can’t do that. That’s not even covered by warranty. Sorry.”