{"ok":true,"nota":{"id":"562ada87341186dc","slug":"mas-de-la-mitad-de-los-papers-academicos-de-2025-muestra-senales-de-te-562ada","titulo":"Más de la mitad de los papers académicos de 2025 muestra señales de texto generado por IA","resumen":"Un ensayo cuenta que una estimación sitúa en 57 por ciento los artículos académicos de 2025 con rastros de modelos de lenguaje, frente al 12 por ciento de 2023. El autor también generó un paper falso con IA para ver qué pasaba al enviarlo.","fuente":"ieee","fuenteNombre":"IEEE Spectrum","color":"#00629b","autor":"Benjamin Skuse","url":"https://spectrum.ieee.org/ai-generated-scientific-papers","imagen":"https://spectrum.ieee.org/media-library/man-in-glasses-at-night-his-face-reflected-in-a-window-beside-him.jpg?id=68515605&width=980","tema":"ciencia","subtemas":["publicación científica","inteligencia artificial","revisión por pares","desinformación académica","integridad científica"],"entidades":["IEEE Spectrum","Nature Communications"],"formato":"dato","entretenimiento":false,"pubTs":"2026-10-27T13:00:03.000Z","indexTs":"2026-10-27T13:32:50.789Z","hash":"63d5c05dfdd10a74d2862adf783de0807b42adf3453fd3f611022e7ba23b1544","tsa":{"time":"Oct 27 13:34:28 2026 GMT","provider":"freetsa.org","sobre":"sello"},"stats":{"views":0,"likes":0,"clicks":0,"dwell":0},"tipo":"normal","registro":"hallazgo","porQueImporta":"La ciencia se basa en que alguien lee, revisa y confía en lo que se publica, y ese filtro se está saturando de texto barato. Cuando el spam llega a las revistas más serias, el riesgo es que las conclusiones que llegan a médicos, docentes y periodistas tengan menos respaldo del que aparentan. Para cualquiera que se informe con estudios, la pregunta es cuánto de lo que lee sobrevive a una revisión real.","quienFirma":"IEEE Spectrum","confianza":"divulgacion","consulta":null},"sello":{"hash":"63d5c05dfdd10a74d2862adf783de0807b42adf3453fd3f611022e7ba23b1544","sello":"35e4ddc649498e3df515abe37260802b09c394475739fbf32e348c487168002d","beacon":{"drand":{"cadena":"8990e7a9aaed2ffed73dbd7092123d6f289930540d7651336225dc172e51b2ce","ronda":6589234,"aleatorio":"28c989e7a1aed4fc51c1fb98e56c6b9badfac1025d9413be7796e49627b5fcd8","tiempo":"2026-10-27T13:34:00.000Z"},"nist":{"cadena":2,"pulso":2001772,"salida":"0E703280D2ECB99CF237148891BE4B05125CA4A9FA93BA99C93DE108693398FC1C1436A917C2E942D3FD99D4FACCDDC40E70589522F4EEDE9B18514294B0BD55","tiempo":"2026-10-27T12:55:00.000Z"},"noAntes":"2026-10-27T13:34:00.000Z"},"verificacion":{"drand":true,"nist":true},"tsa":{"time":"Oct 27 13:34:28 2026 GMT","provider":"freetsa.org","sobre":"sello"},"original":{"titulo":"I Used AI to Fake a Scientific Paper. Here’s What Happened","descripcion":"Science has a spam problem. And the spam filter is getting clogged up. According to all reports, academic publishing is drowning in what has come to be called AI slop: low-quality, mass-produced content generated with AI. By 2025, an estimated 57 percent of all published academic articles exhibited detectable evidence of large language model (LLM) influence, a massive jump from 12 percent in 2023. What’s more, over 12 percent of Nature Communications reviews were found to be completely AI-generated in 2025. Given that for every 40 papers showing statistical evidence of AI usage in 2025, only one formally disclosed it , these figures likely grossly underestimate LLM involvement today. Of cour","pubDate":"Tue, 27 Oct 2026 13:00:03 +0000","url":"https://spectrum.ieee.org/ai-generated-scientific-papers"},"indexado":"2026-10-27T13:32:50.789Z"}}