internet data
1 statements across 1 episodes · 0 bullish · 0 bearish · 1 people on the record · first statement Feb 5, 2026 by Didi Das · across every show →
Everything said about internet data, oldest first
Feb 5, 2026 neutral
Das: Arbitrary LLM scaling fails once internet training data is exhausted
“It turns out that the scaling laws are proportional to the amount of data you can train on. So you can't just arbitrarily make the model super big if the data set is just the internet. Yeah, maybe you can accentuate that with other kinds of like proprietary in…”