The Mexican publisher Almadía has suspended the Spanish promotional tour for David Silva Rojas’s novel Donde el agua trabaja after a detector flagged the text as 99% AI-generated. The controversy erupted when El País employed Pangram, an algorithmic analysis tool, to verify the winner of the Almadía-Ventosa Arrufat prize. Although the contest rules explicitly prohibited artificial intelligence use and Silva Rojas had sworn compliance, the detection result triggered immediate cancellation of events in Barcelona and other cities.
The incident highlights a growing tension between traditional literary integrity and automated verification. Silva Rojas denied the accusations, while editor Guillermo Quijas announced a "detailed review" of the situation. Critics have dubbed these detectors the "VAR of literature," raising concerns about their fallibility. As noted by El Colombiano, no tool is currently infallible; subtle human edits or minor AI assistance can evade detection, yet heavy reliance on algorithms often yields high scores.
The debate extends beyond this single case. Similar scrutiny recently affected Goncourt finalist Thélyson Orélien. Writers like Olga Tokarczuk and Jorge Volpi argue that using AI for research or drafting is acceptable if transparent, whereas others view any machine involvement as a betrayal of artistic doubt. Patricio Pron suggested the surge in AI novels reflects a broader depreciation of literature driven by overproduction. Meanwhile, publishers like Planeta are implementing strict filters, signaling that future awards may function more like tribunals of authorship than merit-based selections.
