In 1923, an editorial cartoonist named HT Webster drew a humorous cartoon for the New York World. This cartoon depicts his fictional 2023 machine that generates ideas and automatically draws them as cartoons. It foreshadowed his recent advances in AI image synthesis a century after he could actually create artwork automatically.
A vintage cartoon is captioned, “In 2023, when all our work is done by electricity.” It depicts a cartoonist standing by his drawing table and planning a social event while the “idea dynamo” generates ideas and the “manga dynamo” renders the artwork.
Interestingly, this division of labor is similar to neural networks today. In fact, he said that in 2023, the “idea dynamo” is likely to be a large language model like GPT-3 (though imperfect), and the “cartoon dynamo” is an image synthesis model like Stable Diffusion. most similar to
In 2014, the blog Paleofuture featured Webster’s work, specifically this cartoon, noting that in the early 1920s, only 35% of Americans had electricity in their homes. Electricity and the devices powered by it represented a radical new way of getting things done. Yesterday, someone on Reddit spotted the cartoon again and it went viral on social media.
Interestingly, despite rapid advances in generative AI techniques over the past two years, image synthesis models are still not very good at line drawings. be careful on Twitter.
But improvements to the AI model that masters hand-drawn cartoons may come soon. And unlike other future predictions of the early 1900s, which involved individual butterfly wings or city-wide air pipe networks, Webster’s prediction seems pretty spot on.
