September 10, 2025Open Access

Leveraging Artificial Intelligence Large Language Models for Writing Clinical Trial Proposals in Dermatology (Preprint)

View Full Paper

Ask AI

Bookmark

Chat GPT o1 was rated the most accurate proposal creator, surpassing all other LLMs and human writers.
Rating results showed LLMs scored proposals nearly 1.9 points higher than human reviewers for accuracy and comprehensiveness.
This analysis used ten LLMs and multiple reviewers to assess their effectiveness in drafting proposals for clinical trials.
While LLMs assist in proposal writing, they cannot fully replace expert input, emphasizing the need for human oversight.

Discussion

No takes yet. Share an insight, caveat, or question.

Comparison of LLMs and human composition reveals strengths and weaknesses in accuracy and comprehensiveness.

Hauptman et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark