But the technology’s capacity for imitation will likely continue to improve: New Yorker staff writer Kyle Chayka observed that ChatGPT was not very effective at mimicking his own writing voice, but the AI startup, Writer, created a bot trained on his own writing to produce text in his voice that, while not perfect, was still “unnervingly effective.

Writers who use AI as a drafting tool might be seen as intentionality remixers, drawing copy from ChatGPT as raw material and weaving purpose into the syntax and language. But so far, the chatbot on its own, or even with the support of savvy prompt engineering, does not excel at this particular task.

But what is general knowledge in a world where virtually any information is freely accessible online?

Their relevance is debatable, too — what’s the value we’re trying to preserve by differentiating human writing from AI? Is it really plagiarism not to cite an AI-generated phrase? Is plagiarism still the crime we think it is?

As we explore new applications for large language models and consider how well they can optimize our communication, AI challenges us to reflect on the qualities we truly value in our prose. How do we measure the caliber of writing, and how well does AI perform?

language is a lens, and our thoughts and perceptions are almost certainly shaped in some way by its conventions and metaphors. It shows us that writing influences our thoughts even as we use it to describe them.

simply abiding by the rules doesn’t make excellent writing — it makes conventional, unremarkable writing, the kind usually found in business reports, policy memos and research articles.

Good character impressions delight us because they illuminate something fundamental about the person they imitate

It’s difficult to determine a text’s quality without considering the context in which it was written.