Skip to content
Modelwire
Subscribe

Engineering journals show 82% surge in AI-like language after ChatGPT launch

Source published ·Modelwire updated

Original coverage: arXiv cs.CL ↗·How Modelwire adds context

The development

A large-scale analysis of 1.8 million engineering articles reveals measurable shifts in writing patterns following ChatGPT's November 2022 release. Researchers tracked vocabulary, style, and structure across 256 journals from 2015 to 2025, finding that language associated with LLM output appeared 82% more frequently than pre-AI trends predicted by 2025, while formal control words declined. This empirical study quantifies how generative AI is reshaping scientific communication itself, offering a concrete baseline for understanding adoption velocity and stylistic drift in knowledge work.

Modelwire’s AI-generated summary of coverage from arXiv cs.CL.

Modelwire analysis

Analyst take

Our AI-generated reading of the wider context and the next developments to watch.

The study doesn't just show that engineers write differently post-ChatGPT; it establishes a quantifiable baseline for how fast stylistic adoption occurs across a knowledge sector. The 82% deviation from pre-AI trends by 2025 suggests adoption velocity is outpacing what historical technology transitions would predict.

This connects directly to Medium's AI-first authoring pivot (November 9) and the Australian universities embedding ChatGPT into curricula (November 5). Both assume writers and students will adopt AI assistance if the friction is low enough. This paper provides the evidence: they already have, measurably, across 1.8 million articles. The stylistic drift also echoes the linguistic bias finding (November 6) that showed how LLM output carries detectable patterns. What's new here is scale and speed. The broader pattern across recent coverage is that AI adoption in knowledge work is no longer aspirational; it's already baked into how people write, learn, and work.

If the same 82% deviation holds when researchers isolate articles from institutions with explicit AI-use policies versus those without, that confirms adoption is driven by institutional enablement rather than individual preference. If the stylistic markers begin reverting or stabilizing in 2026 data, that signals either saturation or a backlash cycle.

This interpretation is generated from the summary above and the archive coverage cited below. Our methodology · Report an error

Coverage behind this analysis

These archive entries ground the connection in our analysis. They are ordered by source publication date, with links to our coverage and the original sources.

  1. ·OpenAI

    Australian universities integrate ChatGPT into core curricula

    Australian universities are embedding ChatGPT into curricula and expanding student access to large language models, signaling a structural shift in how institutions prepare graduates for AI-augmented workforces. This move reflects growing recognition that AI literacy is now foundational rather than specialized, pushing higher education to treat LLM fluency as a core competency alongside traditional subjects.…

    Read Modelwire coverage →Original source ↗

MentionsChatGPT · OpenAI

MW

How this coverage is produced

Modelwire uses AI to generate summaries and context from source headlines, snippets, and selected archive coverage. Automated checks do not verify every claim, and items are not routinely reviewed by a person before publication. Zacaria Solis operates the site. Read the linked source for the full evidence and report errors through our corrections process.

Modelwire summarizes, we don’t republish. arXiv cs.CL originally reported this story as “What Changed in Engineering Writing After Generative AI Chat Assistants: Evidence from 1.8 Million Articles”. The full content lives on arxiv.org. If you’re a publisher and want a different summarization policy for your work, see our takedown page.

More to explore

Five-year study finds LLMs have not yet reshaped scientific publishing

arXiv cs.CL·

Stack Exchange reveals where AI-generated answers slip past moderation

arXiv cs.CL·

OpenAI surveys 1,800 researchers on ChatGPT's role in scientific workflows

OpenAI·
Engineering journals show 82% surge in AI-like language after ChatGPT launch · Modelwire