Large language models (LLMs) learn from the open web, and that reliance should be seen as one of the most important cybersecurity vulnerabilities. No wonder then that malicious actors take them as a target. In the OpEd “Mój chatbot Władimir. Jak Rosjanie zatruwają wielkie modele językowe” (published by the Polish Spider’s Web, co-authored with Rafał Pikuła and M. Szymaniak), Mateusz describes how networks such as Russian “Pravda” flood the internet with low-quality content so that chatbots later repeat Kremlin narratives. Truth drowns in a sea of cheaper information.

This article opens a series of Mateusz’s media appearences in which he explains the architecture behind LLMs and the threats for democratic processes. In OpEd written for Kontakt titled “Dezinformacja w dobie sztucznej inteligencji”, he shows the full spiral: AI creates content, bots amplify it, algorithms spread it, and people then ask chatbots to check it.

In July, PAP (Polish Press Agency) asked him about a chatbot that repeated a fake report of Donald Trump’s death, planted through a Reddit “data poisoning” campaign.

And in August he discussed LLM grooming in “Audycja Międzynarodowa” on Polish Radio Trójka. You can check out the links below.

Spider’s Web OpEd: https://spidersweb.pl/plus/2026/04/jak-rosjanie-zatruwaja-wielkie-modele-jezykowe
Kontakt OpEd: https://magazynkontakt.pl/dezinformacja-w-dobie-sztucznej-inteligencji/
Polska Agencja Prasowa: https://www.pap.pl/aktualnosci/ai-powielila-falszywa-informacje-o-smierci-trumpa-naukowiec-modele-ucza-sie-z-forow
Polskie Radio Program Trzeci: https://www.polskieradio.pl/trojka/audycje/audycja-miedzynarodowa,11613/odcinek/aktualna-sytuacja-w-ukrainie-audycja-agnieszki-lichnerowicz,9e6620f9-5c48-4ada-9206-a5754195b7cd