The image is by traditional-word.
The results of training neural networks with imagery from The Simpsons are demonic. The latest in a long line of multimedia artists who use machine learning to pay homage specifically to The Simpsons was posted on the MediaSynthesis subreddit. The Simpsons hold a special place in the hearts of generative artists. The recent buzz surrounding the work of Brazilian generative artist Hidr who used a combination of artificial intelligence and software to turn beloved characters from the show into creeps.
The Boeing DreamKitten 787 was posted on January 31, 2022.
It gets even weirder. Tim McSmythurs, a Canada-based artificial intelligence researcher, made waves when he trained a model that could mimic anyone's voice. McSmythurs used characters from the long-running animated series into other media and put people in the show that didn't actually guest star, including Donald Trump.
McSmythurs said in an interview that one could come up with an episode of The Simpsons that was voiced by the characters in a believable way.
Wait, there's more! A few years ago, someone trained another neural network to generate a script that was so bad it was called soulless Simon Sam.
Back then, text generation artificial intelligences were not as advanced as they are today, and the characters that came out of them were pretty weird.
Neural networks have come a long way since Sam Simon started unleashing nonsensical hell on the machine learning community. As the technology gets smarter, so will the art it spits.
Scientists are concerned that new artificial intelligence-generated faces are seen as more trustworthy than real ones.
Are you interested in supporting clean energy adoption? At UnderstandSolar.com, you can find out how much money you could save if you switched to solar power. Futurism.com may receive a small commission if you sign up through this link.
All Rights Reserved © 2024
