fr | en

Publications

Research, policy briefs and analyses from CeSIA.

All Report Article Op-Ed Press Release Event

Anderson Mancini, Three wise monkeys, cc-by-2.0

Article Sep 2026

Responses to common objections about the risks of advanced AI

Arthur Grimonpont

Capture d’écran 2026-08-27 à 11.13.22

Article Jul 2026

The OpenAI–Hugging Face incident: a detailed timeline and what it means

Charbel-Raphaël Ségerie & Arthur Grimonpont

harmonizing thresholds article

Article Jul 2026

A methodology for harmonizing risk thresholds across different AI companies

Markov Grey

AI Companies Have No Common Standards for Defining When AI Becomes Dangerous: A concrete proposal

Jul 2026 Markov Grey

Read the article →

Capture d’écran 2026-08-10 à 16.39.04

Article Jul 2026

Charbel-Raphaël Ségerie has been elected to the board of the International Association for Safe & Ethical AI (IASEAI).

What might an international AI treaty look like? A concrete proposal

Jul 2026 Charbel-Raphaël Segerie

Read the article →

Safety tops the agenda of the UN Global Dialogue

Jul 2026 CeSIA — Centre pour la Sécurité de l'IA

Read the article →

CeSIA presenting the HAIP reporting changes at the OECD

Article Jun 2026

Version 2 of the Hiroshima Process is good news for AI transparency

Charbel-Raphaël Ségerie & Jonathan Salter

Screenshot of paper

Article Jun 2026

BELLS-O: an evaluation of 28 AI guardrails from 17 providers

Hadrien Mariaccia

Consultation PPF x CeSIA on cybersec

Article May 2026

CeSIA and the Paris Peace Forum launch a joint consultation on AI safety and cybersecurity research