Have you ever wondered what it would look like if a giant humanoid lobster was photographed playing basketball? 

Or, perhaps, how Pablo Picasso would have painted President Trump?

What about if the January 6 Capitol riot was led by the Minions from “Despicable Me?”

New artificial intelligence technology is allowing the Internet’s inhabitants to create images of all these things and far, far more within seconds.  

Dall-E Mini, a “text-to-image” application that became publicly available in April, has gone viral online during the past week. Users across social media have used the app to create and post uncanny images that can be generated by the app simply by typing any combination of words into the search bar.

Dall-E, whose name is a portmanteau of the painter Salvador Dalí and Pixar’s beloved robot “WALL-E,” was created in 2021 by OpenAI, an artificial intelligence research lab based at San Francisco.  The company’s mission, its website says, “is to ensure that artificial general intelligence benefits all of humanity.”

The A.I. relies on a vast “neural network” that has been trained to associate specific words and phrases with collections of billions of web images.

“We’ve found that it has a diverse set of capabilities,” OpenAI’s 2021 paper explaining the technology says, “including creating anthropomorphized versions of animals and objects, combining unrelated concepts in plausible ways, rendering text, and applying transformations to existing images.”

The results are stunningly accurate depictions of disparate objects and concepts woven together logically.  One of the test cases by Dall-E’s creators was to create “an illustration of a baby daikon radish in a tutu walking a dog.” 

Dalle-E Mini, the version available to the public, is based on the open-source programming and learning capabilities published by OpenAI, but is unaffiliated with the program’s creators.  

It is noticeably less advanced than the project it is based on, relying on less complex code and a smaller database of images.  The result is that the images are often warped in horrifying ways while still clearly conveying the idea that was intended.

Take this nightmarish image of President Biden caught on a “trail cam”:

While still ludicrously impressive, Dall-E Mini’s greatest weakness is its rendering of faces.

According to the app’s programmer, Boris Dayma, the app is much better at “abstract painting” than constructing precise images.

“If you draw a landscape with Dall-E, it’s amazing, because if there’s a small problem with a tree, nobody notices it and the landscape still looks great,” he said, according to Newsweek.  “But if there’s a problem with a face, we notice it. If there’s a small flaw with an eye, we can see it.”

Nevertheless, the app is still capable of rendering a recognizable image of nearly anything, making it a rich vein of endless viral content. It provides users a remarkably simple creative outlet with added bonuses of surreal horror and unintentional comedy.

Dall-E Mini can be found at HuggingFace.co.