The first AI Act obligations for every organization deploying AI apply from 2 August 2026. High-risk systems must conform by 2 December 2027. We tell you exactly what applies to you, close the gaps, and generate the technical evidence regulators and customers will ask for.
Free AI Act Pre-Assessment. Answer ~25 structured questions, get an instant scored snapshot: your risk tier, your top gaps, and your personal obligation timeline. The full self-service portal launches at compliance.cyberfort.lv in July — request early access now and we'll run your pre-assessment with you, free.
Start Free Pre-Assessment →Updated for the May 2026 Digital Omnibus agreement — beware of outdated advice still quoting August 2026 for high-risk systems.
Prohibited AI practices and staff AI-literacy duties (Art. 4, 5) apply since February 2025. GPAI obligations since August 2025. Fines up to €35M or 7% of global turnover.
If your organization uses AI that interacts with people or generates content, disclosure and transparency duties (Art. 50) apply — chatbots, copilots, AI-generated content included.
HR & recruitment, credit scoring, education, critical infrastructure, public services. Full conformity: risk management, data governance, documentation, human oversight, monitoring.
Productized, fast, and priced for decision-making — not open-ended consulting. All prices excl. VAT. Pay online by card or bank link, or order by invoice.
The fastest way to know exactly where you stand. An expert-verified assessment built on our structured, AI Act article-mapped methodology — the same model used in national trust frameworks.
Organization-wide assessment across all your AI systems, with a concrete implementation plan for the August 2026 transparency layer and a board-ready view of the road to December 2027.
Full conformity preparation for Annex III high-risk systems against the 2 December 2027 deadline — with the one thing advisory firms cannot deliver: real adversarial testing as your Article 15 technical evidence.
Compliance is not a one-off project. Continuous post-market monitoring, regulatory watch, and a measurable compliance score that improves quarter over quarter.
Latvia's Kiberdrošības likums (NIS2 transposition) has been in force since September 2024 — cybersecurity manager appointments, self-assessments, and annual security reviews are already mandatory for essential and important entities. Roughly 40% of the underlying controls overlap with AI Act obligations: risk management, logging, incident reporting, documentation, governance. Every Cyberfort compliance bundle delivers a cross-mapped gap matrix covering both regimes in a single assessment — in Latvian or English, with regulator-fluent local delivery.
Law firms write policies. Advisory firms write roadmaps. Cyberfort additionally runs the adversarial tests that prove your AI systems are accurate, robust, and secure — the technical evidence Article 15 demands. Our AI Cybersecurity line (LLM red teaming, guardrails, monitoring) plugs directly into every compliance engagement.