EE
FR Contact

AI agent incidents in France

Every incident in the register with a French connection: the AI was built or run there, the people it affected are there, it happened there, or researchers there found it.

2 incidents 6 sources 103 worldwide

As of 26 Oct 2026, updated every morning.

Days without a reported agent incident in France

341

Last reported 19 Nov 2025Grok posts Holocaust denial in French on X; Paris prosecutors widen X probe

2 incidents

All 103 incidents worldwide
  1. EEAIR-2025-043 Nov 2025 Grok posts Holocaust denial in French on X; Paris prosecutors widen X probe Grok (the @grok reply account on X), xAIFrance French-language post; Paris prosecutors and French ministers acted on it. 91%
  2. EEAIR-2024-009 Oct 2024 Obfuscated Imprompter prompt makes Mistral's Le Chat leak users' personal data Le Chat, Mistral AIProof of conceptFrance Le Chat is run by French AI company Mistral AI. 92%

AI agents and French law

Accurate as of October 2026. The designation, both bills and the liability reform are pending.

The EU AI Act applies in France, but no law has yet designated the authorities that will enforce it. The government’s plan of 9 September 2025 makes the DGCCRF coordinator and gives the CNIL several banned practices and high-risk uses (biometrics, employment), the ACPR credit scoring and insurance pricing, and Arcom some transparency duties. The bill enacting it, passed by the Senate on 18 February 2026, awaits the National Assembly, as does the bill transposing NIS2.

Personal-data breaches posing a risk go to the CNIL within 72 hours. On 20 July 2026 the CNIL confirmed the GDPR already applies to agentic AI. On 13 April 2026 ANSSI’s CERT-FR said autonomous assistants such as OpenClaw should not be deployed on workstations. No liability rule is specific to AI; the Civil Code must take in the new EU Product Liability Directive by 9 December 2026.

Sources

  1. DGE: authorities proposed to enforce the EU AI Act (9 Sep 2025)
  2. National Assembly: EU-law adaptation bill (DDADUE), legislative file
  3. CNIL: notifying a personal data breach
  4. CNIL and CIANum: exploratory note on agentic AI and personal data (20 Jul 2026)
  5. CERT-FR (ANSSI): risks of agentic AI automation tools on workstations (13 Apr 2026)

Deploying agents in France?

The EE Agent Safety Protocol, in development, fits guardrails, evaluations and sandboxing to your agents. Talk with the team.