In a new study designed to test the co-creative capabilities of large language models (LLMs), internet memes created by OpenAI's GPT-4o model were, on average, rated funnier, more creative and ...

"I regret to announce that the meme Turing Test has been passed," said Wharton professor Ethan Mollick of the study. "LLMs produce funnier memes than the average human, as judged by humans.

A new study finds that AI-generated memes are now rated funnier than those made by most humans, signaling a major leap in AI's creative capabilities—though the best human memers still hold the edge.

This, below, ranks very highly—third for creativity and fourth for humor. A meme that scored #3 for “Creativity” in the study, captioned by a human. It happens to be “human-made.” ...

In a study, memes created by OpenAI's GPT-4o model were, on average, rated funnier, more creative and more shareable than those created by humans.