The advent of Generative Artificial Intelligence has profoundly reshaped the landscape of translation. Moving beyond traditional machine translation paradigms, large language models (LLMs) now operate as translation agents capable of producing linguistically fluent and stylistically complex texts. As a result, translation is no longer only a matter of accuracy or adequacy, but also one of style.

As LLMs are adopted for translation tasks, their outputs reveal distinctive linguistic and stylistic patterns. These patterns differ in subtle but consequential ways from those found in both human translation and conventional MT systems. While such differences are often perceived intuitively by readers and practitioners, they remain underexplored from a systematic, research-driven perspective.

This evolving scenario raises a set of pressing questions:
What are the stylistic features of GenAI-produced translations?
How do they differ from those generated by traditional MT systems?
And how do they compare to human translations across genres, languages, and contexts?

The StyGenAI Workshop brings together researchers and practitioners interested in the stylistics of AI-generated translation. The workshop focuses on recurrent stylistic patterns in LLM-based translation, departures from human translation style, and the linguistic, technical, and contextual factors that shape GenAI output. Particular attention is given to variables such as text genre, language pair, prompting strategies, and workflow design.

The workshop provides a forum for interdisciplinary dialogue at the intersection of translation studies, computational linguistics, stylistics, and AI evaluation. Contributions are welcomed from both empirical and conceptual perspectives, as well as from research that bridges academic inquiry and professional practice.