The short version
- Santiago Schnell resigned as Dartmouth College provost following an advisory committee finding that he did not disclose substantial AI assistance in his writing.
- The investigation noted insufficient cooperation from Schnell, who withheld personal account logs but provided access to institutional enterprise accounts showing extensive editing and code review activities.
- Despite the leadership change, Dartmouth continues to promote AI integration, with President Sian Leah Beilock emphasizing the need for transparency and high ethical standards among faculty.
Santiago Schnell has resigned from his position as provost at Dartmouth College after an internal advisory committee determined that he failed to disclose significant use of artificial intelligence in both academic publications and a widely read opinion piece. The resignation marks a pivotal moment for the institution, which historically claims credit for launching the field of artificial intelligence in 1956. The controversy centers not on whether Schnell used AI tools, but on his failure to reveal that usage to publishers who require such transparency, raising broader questions about academic integrity in an era of rapidly evolving technology.
The advisory committee’s investigation concluded that Schnell did not cooperate sufficiently with the inquiry process. While he surrendered his university-issued laptop, he refused to provide logs from his personal large language model accounts related to the writings under scrutiny. He also declined to meet with committee members. However, investigators were able to access records from a Claude enterprise account provided by Dartmouth. These logs revealed that Schnell used the AI tool for revising papers, identifying and fixing bugs in code, and editing or generating figures. The extent of this usage appears to have gone beyond simple grammatical corrections.
Schnell has publicly maintained that he viewed AI as an editorial aid and a discussion partner rather than a ghostwriter. In statements to media outlets, he described using chatbots to refine arguments, check for grammar errors, and improve the clarity of his sentences. He told one journal editor that he did not disclose his AI use because he did not believe this type of assisted editing required formal acknowledgment. He compared his usage to employing a spelling or grammar checker, suggesting he may have underestimated the ethical obligations surrounding such tools in academic publishing.
The committee report challenges Schnell’s characterization of his actions. Investigators noted that his public statements likely understated the nature and scope of his actual AI use. Evidence from the enterprise account shows him prompting the AI to act as a peer reviewer for a paper on chemical dynamics, asking it to generate a referee report based on specific expertise in nonlinear chemical kinetics. He also instructed the AI to make corrections to TeX files and repositories, including generating new figures when necessary. This level of involvement suggests a more active role for the AI than Schnell’s descriptions imply.
Particular concern arose from logs showing Schnell uploading a manuscript titled “Academic Freedom and Catholic Mission” to Claude and asking whether the text contained many signatures of AI writing. He subsequently incorporated changes suggested by the AI into the revised manuscript. This interaction raised questions about whether Schnell was attempting to obscure traces of AI activity in his work. The committee emphasized that the full scope of his usage remains unclear due to his refusal to share personal account data, leaving some aspects of the investigation unresolved.
The incident has drawn attention to the ambiguity surrounding disclosure standards in academic and journalistic contexts. Schnell himself authored a paper in September 2026 titled “Who Wrote This?” which argued against binary classifications of human-written versus AI-written text. Yet his own failure to disclose AI assistance in a Washington Post op-ed and academic papers contradicts the transparency he advocated for. The op-ed, published in August 2026, warned universities about the deeper problems associated with AI cheating, making his subsequent resignation particularly ironic given the subject matter.
Despite the scandal involving its top administrator, Dartmouth College continues to pursue aggressive integration of artificial intelligence into its curriculum and research initiatives. President Sian Leah Beilock has publicly expressed support for expanding AI use at the university, arguing that leaders must model transparency and human judgment. She emphasized that holding oneself to high standards is critical in a fast-changing environment. The university is currently hosting a conference titled “What Must Remain Human?” featuring industry speakers and student hackathons sponsored by major AI companies.
The resolution of Schnell’s tenure underscores the difficulty institutions face in defining acceptable AI use. While Schnell argued his methods were akin to traditional editing tools, the committee found that he regularly exceeded his own public descriptions of usage. The lack of clear guidelines has left faculty and students navigating a gray area where disclosure requirements vary by publisher and discipline. As Dartmouth moves forward, the focus remains on establishing consistent ethical frameworks that balance innovation with accountability.
The resignation serves as a cautionary tale for academic leaders who advocate for technological advancement while struggling to adhere to emerging norms of transparency. Schnell’s case highlights the tension between viewing AI as a benign productivity tool and recognizing its potential to compromise authorship integrity. With personal logs withheld, some details of his usage remain speculative, but the institutional response signals a firm stance on disclosure. The university’s continued embrace of AI suggests that this incident will be treated as an isolated failure of judgment rather than a rejection of the technology itself.
Sources behind this briefing
Go to the original reporting
- Ars Technica↗Dartmouth's provost warned us about AI cheating—but his own AI use cost him his job
