AIToday
Large Language ModelsAI Safety & AlignmentAI Business & IndustrySemafor TechPublished: Aug 27, 2026, 13:00 JST2 min read

AI-written guest columns creep into top US opinion pages

AI-written guest columns creep into top US opinion pages

Key takeaway

  • AI is quietly entering top opinion pages. A Semafor analysis found 10 of 310 guest columns were mostly AI-written.

  • Another 40 were partially AI.

  • The trend follows a controversial AI-written Journal op-ed.

3 Key Points

  1. What happened

    A Semafor analysis found that over the past month, 10 out of 310 guest submissions to The Wall Street Journal, The Washington Post, and The New York Times were at least 80% AI-written, and another 40 were partially AI-generated.

  2. Why it matters

    The debate intensified after billionaire Stanley Druckenmiller's AI-written Journal op-ed was confirmed, with the paper's opinion editor defending AI use as "a fact of modern life." Major outlets like The New York Times and The Washington Post prohibit AI in guest essays, yet flagged pieces still appeared.

  3. What to watch

    The Washington Post and The New York Times were both found to have published AI-written guest columns, including one by a former CISA director and one by Dartmouth provost Santiago Schnell, who admitted using ChatGPT to refine his essay.

Ask the AI about this article →

Context & Analysis

The Semafor analysis reveals that AI-generated content has begun to infiltrate even the most prestigious opinion pages, albeit in small numbers. Over the past month, only a small fraction of guest columns were flagged as predominantly AI, suggesting that human authorship still dominates. However, the presence of partially AI-written pieces indicates a gradual shift that could accelerate as tools become more sophisticated.

The controversy surrounding Stanley Druckenmiller's op-ed highlights a divide in editorial standards. While The Wall Street Journal's opinion editor defends AI use as a modern reality, other outlets like The New York Times and The Washington Post maintain explicit prohibitions. Yet, the analysis found flagged articles in both papers, raising questions about enforcement and the effectiveness of detection tools.

Detection itself is not foolproof. Pangram's claim of a 0.5% false positive rate is contested by some, and Stanford research suggests these tools may be biased against non-native English speakers. Dartmouth provost Santiago Schnell, a native of Venezuela, noted this bias, potentially complicating the interpretation of the analysis results.

FAQ

How did Semafor detect AI-written articles?
Semafor used an AI detector called Pangram, which labels text as AI if more than 80% of it is AI-generated. Pangram claims a 0.5% false positive rate.
What did Stanley Druckenmiller say about using AI?
Druckenmiller said he uses AI for all his writing, just like he uses a calculator for math, noting he moved from an English major to economics.
What are the policies of The New York Times and The Washington Post on AI?
The New York Times prohibits AI use in developing and drafting guest essays. The Washington Post requires guest writers to confirm their submission was not created or manipulated with AI.

Get the latest Large Language Models news every morning

AI-summarized, only the topics you pick — one digest a day via Email, Slack, or Discord.

Free · takes 30 seconds · unsubscribe anytime

Ask AI

Ask AI anything about this article. Q&As are published on this page for other readers too.

Related Articles

Next articleOpenAI custom chip Jalapeño beats Nvidia in tests