Snapchat daily active users

Between 2015 and 2023, AI papers published on arXiv and Snapchat's daily active users grew in near-perfect lockstep, with a 0.97 correlation that prompts one to imagine teams of machine learning researchers pausing between gradient descent updates to send disappearing photographs of their whiteboards. The academic AI community and the ephemeral photo-sharing community would appear to have little in common, except that both grew exponentially during a period when the world was collectively very excited about technology and somewhat unclear about what to do with it. The papers will outlast the snaps. Probably.

AI arXiv submissions grew from roughly 5,000 papers in 2015 to over 50,000 by 2023, reflecting the deep learning revolution, massive increases in research funding, and the democratization of publishing in ML fields. Snapchat's daily active users grew from around 100 million in 2015 to over 375 million by 2023, driven by Gen Z adoption and the platform's ongoing feature evolution. Both metrics ride the same wave of digital platform growth and technology investment enthusiasm from the mid-2010s onward, but the correlation is coincidental — driven by the general expansion of the technology sector rather than any meaningful connection between research output and social media engagement.

A decade of technology optimism and capital flows into digital spaces will make everything tech-adjacent rise together. The correlation describes the era, not the entities.