ComplyEdge Blog — EU AI Act Enforcement & Engineering

August 1, 2026 · Analysis EU AI Act Article 50 Agents

The Cited Decision Standard: EU AI Act runtime enforcement audit trail

From 2 August 2026, Article 50 binds. When an AI agent interacts with a person, the system must disclose what it is and on whose behalf it acts. Here is the obligation, the Commission's guidance on agents, and the Cited Decision Standard for the runtime enforcement audit trail: hash, timestamp, and chain on every decision; article citation when a rule fires; regulator-readable export.

Read article

July 11, 2026 · Engineering Trust Due Diligence

How we hold ComplyEdge to its own standard

We sell AI-Act compliance and we build with AI. The two-minute inspection a technical buyer can run — reproducible benchmark, tamper-evident audit chain, and how we govern our own AI-assisted development — from the public repo, without taking our word for anything.

Read article

July 11, 2026 · Benchmark GPAI EU AI Act

We graded 12 GPAI providers on EU AI Act documentation

Google leads at 83.3%; xAI trails at 11.1%. Industry average 49.6% on public documentation across six GPAI obligation categories — open leaderboard, reproducible from provider evidence files.

Read article

May 18, 2026 · Engineering EU AI Act OPA/Rego

Why we built EU AI Act enforcement on OPA/Rego, not an LLM

Most AI guardrails use an LLM-as-judge pattern. That works as a content filter — it does not work as compliance. We put OPA/Rego on the hot path and the LLM on the long tail. Here is why the order matters.

Read article

EU AI Act enforcement, in the open

The Cited Decision Standard: EU AI Act runtime enforcement audit trail

How we hold ComplyEdge to its own standard

We graded 12 GPAI providers on EU AI Act documentation

Why we built EU AI Act enforcement on OPA/Rego, not an LLM