Impact of AI on professional writing and linguistics

Discussions regarding the impact of artificial intelligence on professional writing have shifted from assessing its resilience compared to coding to examining its effects on linguistic variety.

Initial observations suggested that while AI is rapidly transforming programming through automated code generation, professional writing might remain more resilient. This was attributed to writing being a ‘wicked problem’ that lacks objectively correct solutions, making it difficult for models to replicate the depth, authenticity, and soulful communication of humans.

However, subsequent research published in ‘Nature Human Behaviour’ indicates that the widespread use of Large Language Models (LLMs) is contributing to a decline in linguistic diversity. A study of over 880,000 texts found that while AI tools effectively maintain semantic meaning, they tend to flatten individual style by replacing unique expressions with statistically probable, conventional phrasing. This process has been shown to reduce text complexity variance by as much as 50%. In response, academic programs are beginning to investigate how such assistance impacts critical thinking and epistemic agency.