What AI Will Do to Art
from Aarva Hosted by Spencer Kornhaber
If a machine can generate anything in seconds, where does the actual art happen? In a piece published on 30 June 2026, The Atlantic follows two artists who treat artificial intelligence not as a threat, but as a new kind of paint. While the cultural establishment worries that image generators will flood the internet with cheap, disposable content, Holly Herndon and Mat Dryhurst are busy training synthetic choirs. The article circles a question that sits just past the usual panic: when machines can generate almost anything on command, the hardest part of making art is no longer the execution. It is figuring out what is actually worth making. A profile of artists Holly Herndon and Mat Dryhurst as they prepare an artificial intelligence installation for the Venice Biennale. Following the couple from their Berlin studio to the exhibition opening, the author outlines their argument that creators should actively shape machine learning rather than reject it. It details their process of training models on personal data to test whether algorithms can produce original art instead of automated content. Read at source: The Atlantic For more features and details, visit aarva.app
Embed this episode
No transcript yet.