GoDaddy CEO Says He Won’t Read Work Written by AI

“Because if the person couldn’t take the time to write it, then I definitely shouldn’t take the time to read it,” Bhutani said. He distinguishes using AI to understand a word or phrase from having it do the writing. Developing a clear narrative, he said, helps employees practice what he calls “the art of iteration”—refining their work over time.

Bhutani said he wants to see the observation or hypothesis that first drew an employee to an idea, and why they think it matters. AI, he argued, can draw on what is already on the internet but does not know “what we know deep inside us.”

He looks for signs such as unusually polished prose, which stands out among colleagues who speak English as a second language. Length and tone can also raise his suspicions: some AI-written pieces are too long, while others are short and “punchy and clear.” Researchers at Florida State University have identified words including “delve,” “intricate,” “realm” and “underscore” as possible clues, particularly when repeated. Other patterns include formulaic contrasts, flowery metaphors, perfect grammar, frequent em dashes, uniform structure and conclusions that repeat the introduction.

Those signs are not proof. AI detection tools such as Turnitin can misidentify human writing, and the company advises against relying on its results alone to take action against a writer. Writing and editing are common workplace uses of AI: in Gallup’s May 2026 survey, 51% of workplace AI users said they used it for those purposes.