Dartmouth Provost Santiago Schnell Resigns After AI Writing Investigation

Dartmouth College’s provost, Santiago Schnell, resigned on Wednesday, October 29, 2026, after a faculty committee determined he had used artificial intelligence to meaningfully alter his writing and had not been forthcoming about it. The finding centers on an opinion piece Schnell wrote for the Washington Post on AI cheating in universities, the very subject he now stands accused of misrepresenting in his own prose.

The case began when the news outlet Semafor ran guest columns from various authors through an AI detection filter as part of a broader check on newspaper opinion writing. Schnell’s August 2026 column, titled “Universities are fighting AI cheating. But there’s a deeper problem,” came back flagged as entirely AI-written. Schnell pushed back at the time, saying he had used ChatGPT only to “refine a few arguments” and that the piece reflected his “own thinking.”

That explanation did not hold up to further scrutiny. The student newspaper The Dartmouth fed a range of Schnell’s past writing into an AI detector and found a clear split: material published before ChatGPT’s public release in 2022 read as human-written, while his more recent output showed high rates of likely AI generation.

The irony deepened from there. When The Dartmouth asked Schnell last month to explain his use of AI, he replied in an email that he relies on it for “language refinement, copyediting, improving clarity and organization and making writing processes more efficient.” The paper ran that very statement through an AI detector and found it, too, appeared to be AI-written. Schnell then issued a follow-up statement saying he had “developed the substance and wrote a draft” of the email himself before he “used an AI tool to assist with language and clarity.”

A formal investigation followed, and the faculty committee’s conclusion was blunt: Schnell had used AI to significantly assist his writing and had not been forthcoming about it. Dartmouth President Sian Leah Beilock framed the departure as voluntary. “That is why, as leaders, it is so critical that we model the transparency and human judgment that we expect from our students and scholars,” she wrote in a statement, calling Schnell’s exit “his decision” and wishing him well. Schnell referred questions to his attorneys, who did not respond to a request for comment.

Schnell, a Venezuelan-born mathematical biologist, had held teaching and leadership posts at the University of Notre Dame, the University of Michigan and Indiana University before joining Dartmouth, according to his LinkedIn profile.

Why This Lands Hard at Dartmouth

Dartmouth markets itself as the birthplace of artificial intelligence. The college traces the field to a 1956 summer research gathering of scientists in Hanover, New Hampshire. On Thursday, October 30, a conference marking the 70th anniversary of that project opened under the title “What Must Remain Human?” A provost caught misrepresenting his own use of AI, at the very institution that claims to have invented the field, is the kind of coincidence a satirist would hesitate to write.

What makes this story worth watching is not that a university official used a chatbot to help with an email or an op-ed. Plenty of executives, academics and writers now do that without controversy. The problem was the gap between what Schnell did and what he said he did, repeated across multiple explanations that each failed the same test he was supposedly qualified to judge. A man overseeing academic integrity at a major research university wrote publicly about the dangers of AI-assisted cheating while apparently practicing a milder version of it himself, then offered shifting accounts when caught.

• Schnell resigned as Dartmouth provost, announced Wednesday, October 29, 2026
• A faculty committee found he used AI to “meaningfully alter his writing” and “was not forthcoming about his AI use”
• His August 2026 Washington Post column was flagged as entirely AI-written by Semafor’s filter
• The Dartmouth’s AI-detector review found his pre-2022 writing read as human, his recent writing did not
• His explanatory email to the student paper was itself flagged as AI-written
• Dartmouth’s 70th-anniversary AI conference, “What Must Remain Human?,” opened October 30, 2026

The lesson for any organization grappling with AI policy is not that leaders should avoid these tools. It is that disclosure has to be plain and consistent the first time someone asks. Schnell’s downfall came less from using ChatGPT than from giving three different versions of how much he used it, each one undercut by the same kind of detection software he might have warned a student about. Universities writing AI honesty codes for their students would do well to apply the identical standard to the people signing those codes into policy.

More Info

• Dartmouth provost resigns following accusations he used artificial intelligence for writing | seattletimes.com