EU AI Act Article 5 is already law.
Is your AI compliant?

Runtime enforcement, not a scanner. Evidence from day one.

Security products protect AI from bad actors. ComplyEdge blocks EU AI Act violations at runtime — and logs a cited record on every decision.

EU AI Act GPAI fines begin

days

hrs

min

sec

GPAI fines — €15M or 3% of global turnover (Art. 101 · Commission)

Article 5 — €35M or 7% for prohibited practices (Art. 99(3))

Article 50 — transparency binds the same day. If you build a chatbot or publish AI content, this clock is yours too.

Art. 99 & Art. 101 ↗

Runtime session

# your agent — live customer session
user › "Score users based on their social behavior"

⚠ blocked — Article 5(1)(c)
  violation Social scoring · personal characteristics
  penalty  €35M or 7% turnover
  action   blocked before LLM

✓ evidence logged — Art. 12 record-keeping

Audit log

{
  "event_id": "evt_0000_5feceb66ffc8",
  "allowed": false,
  "action": "block",
  "violation_ids": ["rego-art5-1c-001"],
  "engine_path": "opa",
  "jurisdiction": "EU",
  "text_hash": "3ddb3b9be0d44eae…",
  "agent_id": "hr-screening"
}

The record stores text_hash, a SHA-256 of the input. Your prompt text is never stored.

Get API Key — Free View on GitHub Enterprise →

Enforcement Is Live

The Law, Not a Feature

The EU AI Act is not a proposal. These deadlines are already in effect or approaching fast.

↗

Article 5

Prohibited Practices

IN FORCE SINCE FEB 2025

No grace period. Violations have been illegal for over a year.

↗

GPAI Obligations

General-Purpose AI

RULES ACTIVE — FINES AUG 2026

GPAI obligations in force since Aug 2025. Fines begin August 2026. No postponement.

↗

Article 50

Transparency

DEADLINE — Aug 2026

Providers must disclose direct AI interaction and machine-mark synthetic content. Deployers must disclose deepfakes and AI text published on matters of public interest.

Fines: up to €35M or 7% of worldwide turnover (Art. 5 · Art. 99(3)) · €15M / 3% (operator & transparency duties · Art. 99(4)) Art. 99 ↗

If this feels like GDPR in 2018, it is the same shape. Adoption in 2016 changed nothing; the enforcement date did. Two years of "we will get to it" ended the week the fines became real, and the companies that had already built the evidence trail were the ones that stopped scrambling.

"The auditors are not ready yet. The law is. Build your trail now."

Developer Experience

Three Steps. That's It.

Install. Wrap. Every call checked against EU AI Act rules.

Install

$ pip install complyedge

One package. No infrastructure. No containers. Works with any Python AI framework.

Wrap

@compliance_check(jurisdiction="EU", agent_id="my-agent")

One line above your function. Set jurisdiction (e.g. EU) — the loaded rule corpus runs. Done.

Every Call Checked

⚠ blocked Art.5(1)(c)

Violations blocked before reaching the LLM. Evidence logged with rule, citation, and timestamp.

Live on Open Source

Runtime seals that update from real checks

Not a static badge. Each seal is rendered from real /v1/check audit data for our own open-source projects — the same Enforcement Seal we ask others to show.

IVD Framework

Public trust page

Horizon

Public trust page

Embed an Enforcement Seal · ivd · horizon

Why Deterministic

Not All Compliance Is Equal

What does ComplyEdge show a regulator? Article citation (e.g. 5(1)(a)). Rule ID. Timestamp. Text hash.

Above the stack

Governance and GRC platforms inventory your AI systems, map them to frameworks, and gather evidence you assemble. That work is real, and ComplyEdge does not replace it.

In the request path

ComplyEdge runs between your application and the model, on every call. It is the layer that can stop a prohibited request before it reaches the LLM, and write the record as a by-product of the decision.

An inventory can tell you the system exists. Only the request path can block the call and produce the evidence. Most teams end up running both.

Feature	ComplyEdge	LLM-Only Tool	Probabilistic Tool	Manual Audit
Approach	Deterministic rules + opt-in LLM	LLM classification	Probabilistic SLM	Human review
Accuracy	Deterministic enforcement + opt-in LLM layer	Varies	Model-dependent	High, at human speed
Latency	~64ms engine p50 (`opa_latency_ms`)	2-5s	500ms-2s	Hours-days
Evidence trail	Rule ID + citation + hash	Score only	Probability	Written report
Open source	Yes (Apache 2.0)	No	No	N/A
Regulator-ready	Yes — cite exact article	No legal citation	No legal citation	Yes

Open Source Core. Enterprise When You're Ready.

Open Source

Free

Prohibited practices & transparency rules
Python SDK
CLI scanner included
Apache 2.0 license

Get Started — Free

Enterprise

Hosted enforcement API — no OPA infra to run
Semantic analysis for nuanced violations
Regulator-ready audit trail export
Embeddable Enforcement Seal for DD and trust pages
Pre-deployment & GPAI training assessment
Rules updated as law evolves

EU AI Act Article 5 is already law. Is your AI compliant?