Gabriel  Alcaras is a sociologist and postdoctoral researcher at Sciences Po. He  joined the médialab in September 2024 as part of the “Ecology of LLM  Practices” (EL2MP) project. The project seeks to understand how  generative artificial intelligence - Large Language Models (LLMs), such  as ChatGPT, Gemini, Claude, Llama, etc. - fit into ordinary working  practices. What tasks are delegated to these systems, and with what  effects? How do these artificial intelligences affect the hierarchical  order and moral economies of professional worlds? What is the nature of  the relationship between workers and these programs? To answer these  questions, Gabriel is working with the médialab's design team to develop  innovative qualitative methods that will enable him to follow the use  of LLMs over the long term with great precision.

More broadly,  Gabriel is interested in the production of digital infrastructures, the  use of digital technologies in the world of work, and methodological  issues (qualitative, quantitative, and computational) in the social  sciences. In November 2022, he defended his PhD thesis entitled “Code control. How writing software became industrialized”,  devoted to Git, a hegemonic development infrastructure for collective  code production. He was awarded the Prix du Jeune Auteur by the journal Sociologie du travail, for his 2020 article “Industrial Public Goods. A genesis of the insertion of Free Software in Silicon Valley ”  (First Prize Ex-Æquo). He is also a member of the editorial board of  the online journal RESET - Social Science Research on the Internet.)