Study: AI now creates funnier memes than humans on average

A new study reveals that AI has surpassed humans in creating humorous memes, marking a significant milestone in artificial intelligence’s creative capabilities. The research, to be presented at the 2025 International Conference on Intelligent User Interfaces, shows that AI-generated captions for famous meme templates outperformed human-created ones on average for humor, creativity, and shareability—though humans still produced the most exceptional individual examples. This finding suggests that while AI excels at identifying broadly appealing humor patterns, human creativity remains essential for content that resonates on deeper levels.

The big picture: AI-generated meme captions rated higher on average than human-created ones across multiple measures of quality and appeal.

• Wharton professor Ethan Mollick called this development “passing the meme Turing Test,” referencing Alan Turing’s famous benchmark for machine intelligence.
• The study reveals a nuanced relationship between AI and human creativity, suggesting complementary strengths rather than outright replacement.

Key findings: While AI memes performed better on average, the study found important distinctions in how humans and machines approach creative humor.

• Human creators still produced the funniest individual memes in the study, showing exceptional creative capacity.
• Collaborative human-AI efforts yielded the most creative and shareable content overall.
• Participants using AI assistance generated more meme ideas while expending less effort than those working independently.

Behind the numbers: Researchers attribute AI’s strong performance to its training on vast amounts of internet data.

• The AI model’s access to enormous datasets allows it to identify humor patterns that appeal to broad audiences.
• This pattern recognition capability helps AI generate consistently funny content that connects with the average viewer.

Limitations: The study methodology contained several constraints that could impact the broader applicability of its findings.

• The research featured short meme creation sessions that might not reflect real-world creative processes.
• Potential biases in crowdsourced evaluations could have influenced the comparative ratings.

Why this matters: The study demonstrates AI’s increasing sophistication in understanding and generating culturally relevant humor, a domain previously considered uniquely human.

• As AI continues advancing in creative fields, the findings suggest the most promising future lies in human-AI collaboration rather than replacement.