A Pew Research Center analysis of nearly half a million English-language web pages found that more than a third of pages published after ChatGPT's launch show signs of AI-generated text. The study used the Open Pangram detection tool to analyze texts from the Common Crawl web archive. The findings indicate a sharp rise in AI-written content since late 2022, with commercial .com domains containing AI-generated text roughly ten times more often than .edu or .gov sites.
The analysis revealed that language patterns such as em dashes and Oxford commas have become more common since 2023. Em dashes now appear about twice as often as in 2023, and Oxford comma usage has increased by 63 percent. Certain AI-favored words like 'delve,' 'interplay,' and 'testament' have more than doubled in frequency. Negative parallelisms following the 'it's not just X, it's Y' pattern have nearly tripled, though they remain rare in absolute numbers.
The study also noted that definitions of 'AI text' remain unclear. The spectrum ranges from fully automated content to human drafts polished with AI to texts where a model only contributed a few sentences. Open Pangram and similar models can only make rough assessments about whether a text is AI-generated, and they often misfire. The researchers emphasized that public perception of negative effects often exceeds what the data supports.
Source: thedecoder
