From b06f8996afb9d1feb9616394c7983982085d06b4 Mon Sep 17 00:00:00 2001 From: Bipin Rimal Date: Sat, 21 Mar 2026 02:36:18 +0545 Subject: [PATCH] =?UTF-8?q?fix:=20Split=20Article=2013=20(provider?= =?UTF-8?q?=E2=86=92deployer)=20from=20Article=2050=20(deployer=E2=86=92us?= =?UTF-8?q?er)?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- docs/my-website/docs/eu-ai-act-compliance.md | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-) diff --git a/docs/my-website/docs/eu-ai-act-compliance.md b/docs/my-website/docs/eu-ai-act-compliance.md index bd29ad66caa..d2dc79342e3 100644 --- a/docs/my-website/docs/eu-ai-act-compliance.md +++ b/docs/my-website/docs/eu-ai-act-compliance.md @@ -102,20 +102,27 @@ litellm.failure_callback = ["your_logging_backend"] Connect to a persistent backend (Langfuse, Helicone, or your own database) with a retention policy of at least 6 months. -## Article 13: Transparency +## Article 13: Transparency to deployers -Deployers must inform users that they are interacting with an AI system and provide information about its capabilities and limitations. +Article 13 requires providers of high-risk AI systems to supply deployers with sufficient information — instructions for use, accuracy metrics, known limitations — to operate the system appropriately. This is provider-to-deployer transparency. -LiteLLM's contribution to transparency: -- **Model routing is logged** — you can tell users which model answered their query -- **Cost attribution** — you know which features consume the most AI resources +LiteLLM's contribution to Article 13: +- **Model routing is logged** — deployers can see which model handled each request +- **Cost attribution** — deployers know which features consume the most AI resources - **Fallback chains are visible** — when a primary model fails and a fallback serves the response, this is logged +- **Provider documentation pass-through** — LiteLLM's docs link to each provider's model cards and usage policies + +## Article 50: End-user transparency + +Article 50 requires deployers to inform end users that they are interacting with an AI system. This is deployer-to-user transparency, and it is a separate obligation from Article 13. What you need to add: - User-facing disclosure that AI is involved in generating responses - Documentation of which models are active and their known limitations - Information about how routing decisions are made (cost, latency, quality) +Note: Article 50 applies to chatbots and systems interacting directly with natural persons. It has a separate scope from the "high-risk" designation under Annex III — it applies even to limited-risk systems. + ## Article 14: Human oversight LiteLLM's guardrails feature provides a foundation for human oversight: