This paper explores the “virtuous” potential of generative artificial intelligence in promoting inclusive administrative communication. The main objective is to investigate whether Large Language Models (LLMs) can facilitate the drafting of texts that respect gender language equality, offering concrete support to both those who are required to write inclusive texts and those who wish to adopt gender-equitable language practices. The analysis focuses on a specific text typology, job advertisements drafted by the Swiss public administration, and describes the use of short splitting (as 'experto/a') first in a sample of real advertisements published on the Swiss federal government portal and then in the output generated by nine LLMs using different prompt engineering techniques.

Translated with DeepL.com (free version)