The Wall Street Journal recently published an op-ed written with the help of artificial intelligence. When the choice drew scrutiny, the paper's editorial page editor, Paul Gigot, defended the author's work, calling the use of AI "a fact of modern life."
The defense raises a question with no settled answer: when, if at all, does AI assistance in published commentary cross a line the reader should be told about?
The Free Press took the question outside the newsroom. Rather than adjudicate the Journal's decision, the outlet asked people across professions — teachers, religious leaders, AI enthusiasts and skeptics among them — to describe how the technology is actually shaping their decisions and their work, or how they are declining to use it at all.
According to the outlet's own description of the project, contributors weighed in from different corners of daily life, including Larissa Phillips on farm life and Niall Ferguson on research.
Nellie Bowles, described in the piece as the outlet's "TGIF czar," gave the most direct answer on the record: "I haven't figured out how to use AI for work yet, and I know that makes me a dinosaur, but I just like what I do and how I do it. It's fine. It works."
That is the whole of what the source documents in her voice — a plain admission of non-use, offered without apology.
The record, as far as it has been made public, does not yet include the specific reasoning of the other contributors named. What exists is a framing: a major newspaper's editor normalizing AI-assisted commentary as unremarkable, and a separate outlet inviting individuals to say, in their own words, whether they agree.
Follow the incentive, not the press release. Gigot's phrase — "a fact of modern life" — is doing quiet work. It forecloses the disclosure question before it's asked, treating AI assistance in opinion writing as a settled practice rather than a choice an institution is making on behalf of its readers. That is a defensible position. It is also, plainly, a decision about what a byline is now allowed to mean.
The more interesting data point may be Bowles's. No mandate compelled her either way; she simply hasn't adopted the tool, and says so without hedging. That is what an actual market in technology adoption looks like — not a policy handed down from an editorial page, but individuals weighing a trade-off and choosing, for now, to opt out.
What is true does not need an adjective. The Journal calls AI assistance ordinary; a columnist calls herself a dinosaur for skipping it. Both statements can be accurate at once. The trade-off between speed and authorship, disclosure and convenience, is being settled case by case, writer by writer — not by consensus, and certainly not yet by any rule a reader can rely on.
