GoKawiilStudy flags Claude Opus 5.5's favorite words as AI writing tells
A new study examining Claude Opus 5.5 found the model frequently relies on certain words and phrases, including 'dependable' and 'quietly', that can signal AI-generated text. The research reportedly found that em dashes, long considered a giveaway of AI writing, are not a reliable marker for this model.
GoKawiil's interpretation of the reporting above, not reported fact.
As AI models improve, researchers and readers are looking for subtler linguistic fingerprints to distinguish machine-written text from human writing. This suggests that previous heuristics, like flagging em dashes, may become less useful as models evolve, pushing detection efforts toward tracking specific vocabulary patterns instead.
- Claude Opus 5.5 reportedly overuses words like 'dependable' and 'quietly' in its output.
- Em dash usage was not found to be a reliable AI-writing indicator in this study.
- AI detection methods may need to keep adapting as models change their stylistic patterns.
Source: fastcompany.com, 2026-10-07
Published there as: “New research: Avoid these 3-word phrases if you don’t want to sound like AI”
Read the original report → The summary and analysis above are GoKawiil's own, written from reporting by the source above. Facts and quotes belong to the original publisher.