Large language models are not the problem

TL;DR

The paper argues that concerns about large language models are misplaced, suggesting that the real issue is how the field responds to AI capabilities rather than the models themselves.

Contribution

It challenges the prevailing anxiety about AI by proposing that the problem lies in our approach, not in the models' abilities.

Findings

LLMs can replicate scientific contributions

Anxiety about AI may reflect deeper issues in the field

The focus should shift from models to improving scientific practice

Abstract

If a Large Language Model (LLM) can replicate your scientific contribution, the problem is not the LLM. What does it say about our field that so much of the anxiety about AI comes down to the fear that a machine could do what we do? Perhaps it says we should be doing something better.