The Wall Street Journal has effectively given opinion writers permission to use AI in their drafting process without disclosure. The signal came Monday, when billionaire investor Stanley Druckenmiller published an op-ed in the paper's Opinion section - and then confirmed he'd used AI to help write it.
Druckenmiller's piece, "Let the Bond Market Speak," criticized Treasury Secretary Scott Bessent's leadership. Readers quickly flagged signs of AI involvement, including "Claude-isms" and the "it's not X, but Y" construction that language models favor. AI detection software Pangram also flagged the text.
Druckenmiller didn't deny it. "There's a reason I moved from an English major to being an economics major," he told NOTUS. "I'm not embarrassed by it… I write everything using AI now for the same reason I use a calculator when I do math problems."
He added: "My name is on the piece. It's my message."
The WSJ's position: intent over process
More significant than Druckenmiller's admission was the Journal's response. Opinion editor Paul Gigot told Semafor that AI is a "fact of modern life" and that the paper's standard is whether a piece reflects the author's genuine argument - not how it was drafted.
"People will use it to assist in their work and their writing, including with research, checking grammar, editing and more. The question for us is whether what we publish from contributors reflects an author's original argument, and if the author has the standing and credibility to make it," Gigot said. "In Stan Druckenmiller's case, we have had a relationship with him for many years, and nobody can doubt that his op-ed is his genuine opinion."
That stance puts the WSJ at odds with other major outlets. The Financial Times recently issued a correction after Harvard professor Ricardo Hausmann failed to disclose that AI was used to condense a longer draft of his column - a practice the FT's editorial code explicitly prohibits. The New York Times, after several AI-related controversies involving freelancers, told contributors that "all writing and visuals that freelancers submit to The Times must be the product of human creativity and craft."
The WSJ's approach raises a question the industry will now wrestle with: where does AI assistance end and authorship begin? As Semafor's Max Tani put it, the decision is a "precedent-setting moment in news media."
What this means for writers
For working writers, the WSJ's policy creates a concrete professional benchmark: the paper now treats AI-assisted drafting as acceptable for opinion contributors, provided the argument is genuinely the author's. That's a meaningful distinction from outlets like the FT and NYT, which require human-only drafting for freelancers.
If you write for publication, the practical takeaway is to know your outlet's policy before you use Claude or any other AI tool in your process. The WSJ's position doesn't extend to its own staff, whose use of generative AI is governed by internal policies. And even at the Journal, the standard is that AI can assist - not replace - the author's own thinking. The line between "helping put pen to paper" and "shaping the opinion itself" is exactly where the next disputes will happen.
Druckenmiller's defense - that the calculator analogy applies to writing - will likely become more common as AI for Writers tools improve. But as the FT's correction shows, disclosure expectations vary sharply by publication. When in doubt, ask your editor before you submit.
Your membership also unlocks:
