Regulation Updates
Digital Omnibus Explained: Your New EU AI Act Timeline
The Digital Omnibus deferred the EU AI Act's high-risk deadline to 2 December 2027. Here's what changed, what didn't move, and what it means for your compliance plan.
High-risk rules now apply from 2 Dec 2027
EU AI Act compliance, automated.
Classify, document, and monitor your AI systems before the EU AI Act's high-risk obligations take effect.
Hosted in the EU
Built in Berlin
The problem
Documentation goes stale. Obligations don't.
Most companies treat AI Act compliance as a one-time project. The regulation is ongoing — and the gap between a snapshot and a living system is where the risk lives.
A snapshot, not a system
A consultant maps your systems, writes the technical file, and invoices — €50,000 to €150,000 for one high-risk system. Then a model gets retrained or a vendor gets swapped, and the file is quietly wrong. Nobody finds out until an auditor asks, or a customer’s procurement team does.
Enterprise GRC assumes a team you don't have
ServiceNow and IBM OpenPages do handle continuous compliance — with six-figure budgets, months of implementation, and someone to run it.
The deadline moved; the work didn't
Article 50 transparency applies from 2 August 2026, and grandfathering past December 2027 turns on whether a system has been substantially modified.
The solution
Everything you need to comply with the EU AI Act
Three integrated modules covering the full compliance lifecycle — from initial classification through ongoing monitoring.
Classify
Automatically classify your AI systems under EU AI Act Annex III with our guided assessment engine.
Document
Generate complete Annex IV technical documentation in minutes with AI-assisted content generation.
Monitor
Regulatory-update monitoring keeps you current as the rules move — plus drift detection and incident logging. When dates or standards change, affected systems are flagged automatically.
Covered for your role — provider, deployer, or both
Your obligations under the EU AI Act depend on whether you build an AI system or put one to use. Aikraft covers both sides, and makes clear which is which.
If you're a provider
You build or place the AI system on the market
- Annex IV technical documentation
- Risk-management and data-governance records
- Conformity-assessment evidence and EU database registration
- Post-market monitoring plans
If you're a deployer
You put a third-party AI system into use
- Fundamental Rights Impact Assessment (FRIA)
- Human-oversight assignment and records
- Operational logging and ongoing monitoring
- Serious-incident reporting workflows
See it in action
From setup to audit-ready in minutes
How it works
From zero to compliant in four steps
-
Register your AI systems
Describe each AI system you develop or deploy — name it, tag it by domain, and categorize its primary function. Aikraft supports bulk import via CSV or direct integration with your asset register. Every system gets a permanent audit trail from day one.
-
Get classified
Our guided assessment engine walks you through the Annex III criteria in plain language — no legal expertise required. Answer 10–15 questions about your system's purpose, autonomy, data use, and impact. Aikraft returns a risk tier (Unacceptable, High, Limited, or Minimal) with full reasoning you can share with regulators.
-
Generate documentation
For high-risk systems, Aikraft's AI copilot generates a complete Annex IV technical documentation package — general description, data governance records, testing methodology, human oversight provisions, and post-market monitoring plan. What takes consultants six weeks takes Aikraft minutes. You review, edit, and export audit-ready PDFs.
-
Monitor continuously
Compliance is not a project — it's an ongoing state. Aikraft monitors your classified systems for data drift, logs incidents with regulatory-grade audit trails, and alerts you when EU AI Act guidance changes in ways that affect your documented systems. Dashboards surface risk in real time. Auditors get read-only access via secure links.
Why Aikraft
Purpose-built for mid-market EU AI Act compliance
See how Aikraft compares to the alternatives.
| Feature | Aikraft Berlin, 2026 | Enterprise GRC e.g. ServiceNow | Purpose-built AI Gov e.g. Credo AI | Manual / Consultants DIY + advisory fees |
|---|---|---|---|---|
| Pricing | €199–999/mo | €50,000+/yr | €30,000+/yr | €100,000+ project |
| Setup time | Same day | 6+ months | 3+ months | 4+ months |
| EU data residency | ✓ | ✓ | Not publicly listed | ✗ |
| Self-serve signup | ✓ | ✗ | ✗ | ✗ |
| Aikraft is built for mid-market EU teams | ✓ | ✗ | ✗ | ✗ |
| Auto-documentation Annex IV compliant | ✓ | ✓ | ✓ | ✗ |
| Continuous monitoring Article 72 | ✓ | ✓ | ✓ | ✗ |
Competitor pricing and setup times are based on publicly available information as of July 2026. Where a vendor publishes indicative pricing or a range, the cell links to the source. Results may vary.
Built on trust, built for Europe
Built in Berlin
German engineering, EU values
Hosted in the EU
Your core data stays in the EU
Security built in
Encryption, access controls, and audit trails
GDPR Native
Privacy by design
From the blog
View all posts → Latest EU AI Act insights
Guides
FRIA: When and How to Document
High-risk AI deployers often need a Fundamental Rights Impact Assessment. This guide covers typical triggers, structure, and how FRIA connects to Annex IV and your governance stack.
Regulation Updates
Post-Market Monitoring: What Deployers Must Do
Post-market monitoring under the EU AI Act is not a one-off audit. Here is how Article 72 translates into ongoing logs, reviews, and updates for high-risk AI in production.