All articles tagged with #pangram
Dartmouth Provost Santiago Schnell faces backlash after student reporters found that his 2026 publications were largely AI-generated. Schnell claims he used AI only for editing, but critics argue this violates academic norms and undermines his public stance on AI in higher education.
An investigation by The Dartmouth found that 96% of Provost Santiago Schnell's 2026 publications are flagged as AI-written by the Pangram detector, which has a near-zero error rate. While Schnell claims AI is used only for editing, his pre-2022 works were flagged as 100% human-written, highlighting a sharp contrast in his recent output.
Pangram, a Brooklyn-based startup, has become the dominant tool for detecting AI-generated text in publishing, despite significant criticism regarding its accuracy, potential bias against non-native and neurodivergent writers, and controversial public accusations that have led to canceled book deals.
Pangram’s AI-detector study finds LinkedIn is unusually saturated with AI-created content: 41% of longform posts and 30% of shortform posts are flagged as fully AI-generated. By comparison, Medium longform is 31% AI-generated, X 29% longform AI, Reddit 13% longform, and Substack 10% longform (12% shortform). The data come from Pangram’s Chrome-extension scans that flag AI-generation, highlighting authenticity concerns on LinkedIn and the broader social-media landscape.
The New York Times' Spelling Bee forum provides hints and discussions for the January 6, 2024 puzzle. The puzzle features a grid with the center letter 'r' and surrounding letters 'b', 'c', 'i', 'o', 'p', and 't'. There are 39 words to find, totaling 147 points, including one pangram. The forum also includes a glossary for common terms used in Spelling Bee discussions, assisting players who might feel stuck on today's challenge.