What already applies today?

The prohibitions in Article 5 have applied since February 2025. Since August 2026 the transparency obligations in Article 50 apply: people must be able to tell that they are talking to an AI, and AI-generated content must be labelled. This holds regardless of risk class; for systems already on the market beforehand it applies from December 2026. The obligations for high-risk systems were postponed to December 2027 by the Digital Omnibus.

What is a high-risk AI system?

In simple terms: AI that helps decide things with significant consequences for people, for example in hiring, lending or HR. For such systems the regulation requires documentation, risk assessment and human oversight.

Does this affect us if we only use tools like ChatGPT?

Transparency obligations still apply, for example employees and customers must be able to tell where AI is involved. We map out everywhere AI runs in your business and assign each use to a risk class.

EU AI Act Readiness

The EU AI Act applies in stages; the transparency obligations have applied since August 2026. Are you prepared?

We guide your company safely through the EU AI Act. With pragmatic risk classification, technical expertise and no bureaucratic overhead.

Get in touch

The risk

AI in productive use, but without legal classification or documentation.

Shadow AI with no central inventory
Unclear risk class of the tools in use
Missing legal documentation

Fines of up to

€15 million

or 3% of global annual turnover.

Our mandate

AI inventory & risk classification

We assess the status quo of your systems and classify them precisely into the risk classes of the AI Act. As developers we evaluate your tools directly at code and infrastructure level, not just against generic checklists.

AI competence & training (Art. 4 AI Act)

Since February 2025 Article 4 has required companies to support the development of AI competence within the team. The Digital Omnibus softened this duty, but did not remove it. We teach your staff the legal duties, roles and safe everyday use, hands-on.

Compliance audit & roadmap

A structured comparison of your processes against the legal requirements. You get a prioritised roadmap to close regulatory gaps one by one.

Self-assessment

How well does your company meet the EU AI Act?

Answer 7 targeted questions and get an immediate first assessment plus concrete recommendations for action.

Start the free risk check →

The facts

In force

Transparency obligations (Art. 50) have applied since 2 August 2026

€15 million

Maximum fine or 3% of global turnover

High

Incalculable liability and reputational risk if ignored

Act now

Tick off AI regulation. Free your mind for innovation.

We accompany your company from the first inventory to full regulatory compliance. You get a clear roadmap and a team that flags the gaps and closes them technically.

Get in touch

Why webse.at

We build it ourselves

Risk classification requires deep technological understanding. Because we build AI systems and custom software ourselves, we know exactly how data flows and what happens under the hood.

One dedicated contact

Inventory, training, audit and technical implementation come from a single source. No friction between lawyers without tech know-how and developers without a legal focus.

Data sovereignty in practice

Our own systems run exclusively on EU servers and are fully GDPR-compliant. What we recommend to clients, we live in our own practice every day.

We implement it technically

We show you where the gaps in the system are and close them too. Cleanly integrated, fully documented and audit-proof.

Good to know

We handle the technical implementation, not the legal advice.

For legal assessments we work with specialised law firms. Our strength: inventory, risk classification and audit-proof documentation.

FAQ

What companies ask us about the EU AI Act

What already applies today?: The prohibitions in Article 5 have applied since February 2025. Since August 2026 the transparency obligations in Article 50 apply: people must be able to tell that they are talking to an AI, and AI-generated content must be labelled. This holds regardless of risk class; for systems already on the market beforehand it applies from December 2026. The obligations for high-risk systems were postponed to December 2027 by the Digital Omnibus.
What is a high-risk AI system?: In simple terms: AI that helps decide things with significant consequences for people, for example in hiring, lending or HR. For such systems the regulation requires documentation, risk assessment and human oversight.
Does this affect us if we only use tools like ChatGPT?: Transparency obligations still apply, for example employees and customers must be able to tell where AI is involved. We map out everywhere AI runs in your business and assign each use to a risk class.