Interconnects
The state of the art in open models, RLHF and post-training, from the Allen Institute's Nathan Lambert — the inside view on how frontier models are made. No ads.
A sort of review of the benchmarking industry and what's missing.
US open weights are back!
Open-weights, ideology, and acknowledging trade-offs.
The expanded form of a testimony I prepared for Congress.
An “AI moderate’s” view on recent events and the trajectory of frontier models.
How to get up to speed on open models and their implications.
Some quick notes on a truly weird week.
We’re <5 years into a compounding revolution which could take a century, and how the AI industry should manage this.
