Dartmouth Provost’s “AI Dependency” Sparks Backlash

As leaders of Dartmouth College embrace artificial intelligence as part of their institution’s new value proposition, students and faculty in Hanover are pushing back against a perceived double standard in the college’s AI use policies.

In August, Dartmouth Provost Santiago Schnell published an editorial in The Washington Post urging  universities to offer degrees that “distinguish what students can do independently from what they can accomplish with AI.” Now, revelations that he used AI tools to generate that essay and other writings are fueling an investigation into Schnell’s work and demands for his firing. 

A few weeks later, Semafor published an investigation that identified Schnell’s column as one of 10 guest essays published last month by the Post, The Wall Street Journal and The New York Times that the AI detector Pangram labeled as at least 80 percent AI-generated. “I wrote this essay based on my own thinking. Then, I used ChatGPT to refine a few arguments, check for grammatical errors, and submit it to The Washington Post,” Schnell told Semafor in reference to his article “Universities are fighting AI cheating. But there’s a deeper problem,” which Pangram said was 100 percent AI-generated.

That prompted the student newspaper The Dartmouth to conduct its own investigation, also using Pangram, which has a “near-zero error rate,” according to a 2025 audit by the University of Chicago. According to the investigation the paper published last week, Pangram determined that Schnell’s work was 96 percent “AI-written” across the nine professional articles he published this year, including the Post essay and two academic papers about mathematics, Schnell’s field of study. And in communication with the newspaper about the piece, Schnell also used AI to generate emails, according to The Dartmouth’s analysis.

Schnell defended his work to the student newspaper, stating that while he views large language models as “assistive tools,” they “do not identify the problem, design the research project, guarantee the validity of research results, interpretations, citations, conclusions or substitute for my research, intellectual judgment or authorship.” He also said that the “characterization of my Washington Post essay as ‘100% AI-written’ is not accurate.”

Schnell is not the only provost who regularly uses generative AI tools. According to Inside Higher Ed’s 2026 Survey of College and University Chief Academic Officers, 65 percent said they use AI to draft communications to faculty, staff or students, and 71 percent said they use AI in their work for any purpose at least weekly.

In addition, experts have questioned the accuracy of AI-detection software, which can sometimes return false positives. Some accused students have even sued their universities and won, and many institutions have discouraged faculty from relying on such programs to identify possible cheaters.

‘AI Dependency’?

Nonetheless, some Dartmouth students and faculty say Schnell’s actions violate the college’s own honor code—which requires students to acknowledge if they use generative AI tools to complete their coursework—and are grounds for termination.

“How do we ask our students and colleagues to follow the rules when the academic leader of Dartmouth is allowed to break them?” reads a petition written by Dartmouth’s chapter of the American Association of University Professors. “We call upon you to restore honor and integrity to our institution and repair public trust by removing Professor Schnell from his position as Provost.”

A provost “should be the very embodiment of our rigorous academic standards,” Eli Moyse, a government major, wrote in an editorial for The Dartmouth last week, in which he also called for Schnell’s firing in light of his “AI dependency.”

“I understand how Provost Schnell got to this point. His use of AI is reflective of trends affecting our collective mind. I have seen many of my own peers fall down the same path of increasingly using AI because it’s addictingly easy,” Moyse wrote. “Beyond the Provost Schnell incident, though, lies a deeper reality: Dartmouth’s administration is focusing too much on AI.”

The college, which lays claim to being the birthplace of AI, has made a public push for higher education to embrace generative AI in the classroom.

Earlier this month, Dartmouth President Sian Beilock wrote a guest essay in The Atlantic calling for more AI at the college and warning that “universities that hide from the technology will merely secure their irrelevance.” That declaration followed Dartmouth’s announcement last year of its partnership with Anthropic and Amazon Web Services to continue a “legacy of powerful technological integration at Dartmouth” and ensure “future leaders are equipped to work collaboratively and purposefully with emerging technologies.”

But according to Moyse—who is among the millions of college students bombarded with gloomy predictions and emerging data about AI gobbling up entry-level jobs—it’s all “starting to feel like a top-down preoccupation with the buzz and PR potential of AI is coloring every major choice made by administrators, and Provost Schnell’s use of AI is the natural consequence of this push.”

‘Messy’ Business

Late last week, Beilock responded to the mounting backlash against Schnell. On Friday, she announced that an independent advisory committee composed of faculty and alumni will review “how Santiago has used AI, measured against the criteria of the publications to which he submitted his work.”

But Beilock’s statement also acknowledged that the anger directed at Schnell speaks to bigger issues facing colleges and universities across the country, which are trying to balance academic integrity with pressure to integrate AI into the curriculum.

“Leading on something this new and fluid is messy. There will be missteps, and that’s part of how an institution learns,” she wrote. “The questions this moment raises go well beyond any one person. It is more clear than ever that we need role-appropriate rules and policies for this technology.”

That same day, Schnell published an editorial in The Dartmouth. “I understand that my use of AI in writing is raising important questions about the standards of scholarly and professional work,” he wrote. “In retrospect, I now realize I should always explicitly disclose use of AI in my public writing and be attentive to the expectations of AI use set by publications.” (The Washington Post requires guest editorial writers to confirm their submissions are “not created or manipulated with artificial intelligence or editing software,” according to Semafor’s reporting.)

But as AI use becomes more common, such disclosures may not always be necessary, said Tawnya Means, an education technology consultant and former university administrator. She told Inside Higher Ed that the controversy at Dartmouth is highlighting a need for “meaningful distinctions between assistance and substitution,” when AI is used to generate text.

“Authorship has never required intellectual isolation. Our thinking develops through reading, conversation, feedback, and connections among ideas. AI can participate in that process, although its contribution can also extend much further … ‘AI was used’ tells us very little about the author’s contribution,” she wrote in an email. “Every interaction with AI should not automatically create a disclosure obligation, any more than every consultation with a dictionary or conversation with a colleague does. Specific borrowed ideas still require attribution, and substantial contributions warrant scrutiny.”

Figuring that out will create an “opportunity for higher education … to become much clearer about the human capabilities we want to strengthen, both independently and in interaction with AI,” she added. “The finished response matters, but so does the understanding and discernment that remain with the person who produced it.”