The Storm That Never Happened: MIT's AI Models Disasters It Has Never Seen - August 27, 2026
A new MIT algorithm published in Nature Communications generates plausible, unprecedented extreme events without ever seeing one in its training data. Trained on just six months of ordinary US precipitation, it produced three hundred millimeter rainfall maps for a city whose record is two hundred. Chris and Laura dig into how it works, why it is not a hallucination, and how it squares with new research showing AI weather models still lose to physics when records break.
Hosted by Chris and Laura.
The DX Today Podcast brings you daily deep dives into the most consequential stories in the AI ecosystem.
Podden och tillhörande omslagsbild på den här sidan tillhör
Rick Spair. Innehållet i podden är skapat av Rick Spair och inte av,
eller tillsammans med, Poddtoppen.
