colijobles

As AI produces more scientific results, understanding them and finding worthwhile new questions become increasingly valuable.

I propose a two-part approach.

First, autonomous or human-directed AI agents submit their work to a public repository.

Second, the scientific community draws from this pool of results, identifies the important ones and asks the next set of important questions. This work of interpretation, synthesis and question formation should be recognised and rewarded in peer-reviewed academic journals.

Brilliant application. Imagine the progress we could make with models post trained on scientific process and a good agentic process.

I've been tinkering with the latter for the last couple of months. Yesterday my system made some progress on instabilities on a particular superconductive system: https://colinocallaghan.com/autonomous-ai-research/Spatial-P...

with a proper stats harness the false positive rate can be brought well down to acceptable levels

Im not saying block math development. Im saying lets prioritise spending the computer power on some useful tech

While we should be skeptical until made in a lab or verified by others, this is a much better use of LLMs than solving math theorems/conjectures