Back to today’s briefing

Story perspectives

Study Reveals 13.5% of Papers Use Language Models

7/19/2025

45 10

Summary

Researchers Embrace LLMs

A groundbreaking study of over 15 million biomedical abstracts found that 13.5% of 2024 papers now rely on large language models (LLMs). This trend towards more elaborate language has sparked worries about accuracy, plagiarism, and a troubling uniformity in scientific writing, threatening the integrity of research quality.

Summary

ChatGPT Outdebates Humans

A groundbreaking study in Nature Human Behaviour reveals that ChatGPT can outshine human debaters by leveraging basic personal information to tailor its arguments. This remarkable persuasive ability raises alarms about the potential for AI manipulation in political and social arenas, underscoring the urgent need to address its growing influence on public opinion.

1 of 2

Researchers Embrace LLMs

A groundbreaking study of over 15 million biomedical abstracts found that 13.5% of 2024 papers now rely on large language models (LLMs). This trend towards more elaborate language has sparked worries about accuracy, plagiarism, and a troubling uniformity in scientific writing, threatening the integrity of research quality.

1 / 2