langchain - ✅(Solved) Fix docs: Add EU AI Act compliance guide for LangChain deployers [1 pull requests, 3 comments, 2 participants]

Official PRs (…)
ON THIS PAGE

Recommended Tools

×6

Utilities matched from this issue’s tags and category — try them while you read without losing context.

GitHub issue graph ai analysis

Paste a GitHub issue URL. We fetch that issue, discover linked issues from bodies/comments/timeline, collect linked pull requests, and produce a structured English report.

The report is written in English Markdown for sharing and archival.

Helpful · Quick feedback

Loading…
GitHub stats
langchain-ai/langchain#36173Fetched 2026-04-08 01:17:01
View on GitHub
Comments
3
Participants
2
Timeline
6
Reactions
0
Participants
Timeline (top)
commented ×3closed ×1cross-referenced ×1labeled ×1

LangChain orchestrates LLM calls, agent actions, and retriever pipelines — placing it in scope for EU AI Act compliance when used in high-risk contexts. There is currently no documentation helping deployers understand their regulatory obligations.

Root Cause

LangChain orchestrates LLM calls, agent actions, and retriever pipelines — placing it in scope for EU AI Act compliance when used in high-risk contexts. There is currently no documentation helping deployers understand their regulatory obligations.

Fix Action

Fixed

PR fix notes

PR #36174: docs: EU AI Act compliance guide for LangChain deployers

Description (problem / solution / changelog)

Fixes #36173

Summary

  • EU AI Act compliance guide mapping LangChain's callback system and agent architecture to Articles 12, 13, 14, 50 and GDPR
  • Scope check (Annex III), provider/deployer role clarification, data flow diagram, practical code examples using BaseCallbackHandler
  • References AI Trace Auditor for automated compliance scanning

What's covered

  • Provider vs Deployer — Article 3(3) vs 3(4) role distinction
  • Annex III scope check — is your deployment high-risk?
  • Article 12 — Record-keeping via LangChain callbacks
  • Article 13 — Transparency documentation
  • Article 14 — Human oversight (natural persons, not automated controls)
  • Article 50 — User-facing AI disclosure (50(1) deployer, 50(2) provider)
  • GDPR — Cross-border transfers, DPAs

Developed across 10 open-source AI framework compliance guides. Automated review (Greptile on LiteLLM, Gemini on vLLM) caught and corrected Article 14, provider/deployer roles, Article 26(5) citations, and streaming middleware — all fixes incorporated.

🤖 Generated with Claude Code

Changed files

  • docs/docs/guides/eu-ai-act-compliance.md (added, +427/-0)
RAW_BUFFERClick to expand / collapse

Description

LangChain orchestrates LLM calls, agent actions, and retriever pipelines — placing it in scope for EU AI Act compliance when used in high-risk contexts. There is currently no documentation helping deployers understand their regulatory obligations.

Proposal

Add a compliance guide to the docs covering:

  • Scope check: Is your LangChain deployment high-risk under Annex III?
  • Provider vs deployer roles: Article 3(3) vs 3(4) — different obligations
  • Article 12 (Record-keeping): How LangChain's callback system maps to requirements
  • Article 13 (Transparency): Provider→deployer documentation obligations
  • Article 14 (Human oversight): What deployers must build (natural persons interpreting, rejecting, halting — not automated controls)
  • Article 50 (User disclosure): Applies even to non-high-risk chatbots
  • GDPR: Cross-border transfer considerations per provider

Context

I have a draft ready (was submitted as PR #36171, which was auto-closed for missing an issue link). The guide was developed alongside compliance guides for 9 other AI frameworks (LiteLLM, Haystack, CrewAI, Dify, LlamaIndex, Ollama, vLLM, Langfuse, n8n), with automated code review (Greptile, Gemini) catching and correcting issues around Article 14, provider/deployer distinctions, and citation accuracy.

Generated with assistance from AI Trace Auditor — open-source EU AI Act compliance scanning (Apache 2.0).

Happy to revise based on maintainer feedback on scope, location, and structure.

extent analysis

Fix Plan

To address the lack of documentation on EU AI Act compliance for LangChain deployers, we will create a comprehensive compliance guide.

Steps:

  • Add a new section to the documentation covering the scope check, provider vs deployer roles, and obligations under Articles 12, 13, 14, and 50.
  • Include information on GDPR cross-border transfer considerations.
  • Use the draft guide from PR #36171 as a starting point and revise based on maintainer feedback.

Example Code Snippet (Markdown):

### EU AI Act Compliance Guide
#### Scope Check
To determine if your LangChain deployment is high-risk under Annex III, consider the following factors:

* Purpose of the deployment
* Potential impact on individuals or society

#### Provider vs Deployer Roles
Under Article 3(3) and 3(4), providers and deployers have different obligations:
```markdown
| Role | Obligations |
| --- | --- |
| Provider | Article 12 (Record-keeping), Article 13 (Transparency) |
| Deployer | Article 14 (Human oversight), Article 50 (User disclosure) |

Article 12 (Record-keeping)

LangChain's callback system can be used to meet record-keeping requirements.

Article 13 (Transparency)

Providers must document their AI systems and provide this information to deployers.

Article 14 (Human oversight)

Deployers must implement human oversight mechanisms, including:

  • Natural persons interpreting output
  • Rejecting or halting the system as needed

Article 50 (User disclosure)

Deployers must disclose the use of AI systems to users, even for non-high-risk chatbots.

GDPR

Consider cross-border transfer requirements when using LangChain with providers in different regions.


### Verification
To verify that the fix worked, check that the compliance guide is:
* Added to the documentation
* Easily accessible to deployers
* Accurate and comprehensive in covering EU AI Act compliance obligations

### Extra Tips
* Regularly review and update the compliance guide to ensure it remains accurate and relevant.
* Consider using automated tools, such as AI Trace Auditor, to scan for compliance issues and ensure citation accuracy.

Vote matrix · Quick signals

Works
Did the solution work? Tap to confirm.
Easy Fix
Was it a quick fix?
Time Saver
Did it save you time?
Blocking
Was it severely blocking?
Common Issue
Are others likely hitting this too?
Flaky / Intermittent
Is it intermittent?
Verified / Reproducible
Can you reproduce it reliably?
Loading…

Still need to ship something?

×6

Another batch ranked right after the header list — different links, same matching logic.

Back to top recommendations

TRENDING

langchain - ✅(Solved) Fix docs: Add EU AI Act compliance guide for LangChain deployers [1 pull requests, 3 comments, 2 participants]