What does AI compliance cost? Factors, process and funding options

01 — Price

What it costs

Price

Individual based on effort

Typical duration

3 to 5 weeks

For assessment, classification, and guideline. A fixed price can only be quoted responsibly after the initial meeting — send us a brief description of your project, and we'll send you a proposal that matches the scope and effort.

02 — Factors

What drives the price

• Scope and number of integrations. Every additional system that needs to participate costs effort — regardless of how simple the core is.
• State of your starting point. Clean data, existing content and documented processes significantly reduce effort. Chaos is the most expensive cost factor.
• How much new material needs to be created. What you can contribute doesn't need to be invented by anyone. This is the biggest lever you have in your own hands.
• Requirements for design and customization. A solid standard system costs a fraction of a custom solution — and is sufficient more often than vendors admit.
• Operating model. One-time project costs or ongoing support: calculated over three years, this often gives a different picture than the quote comparison.

03 — Funding

Subsidies in Austria

For digitalization projects, there are several funding programs in Austria, such as KMU.DIGITAL or the aws digitalization funding. Whether and to what extent one of these is currently relevant for your project, we'll review with you individually. Current conditions: kmudigital.at, aws.at.

Which program is relevant, how high the subsidy currently is, and whether the budget for the current funding period is still open changes regularly — plus there are regional funding programs that vary by federal state. Only the information from the funding agencies themselves is reliable, not a fixed percentage stated here.

The most common mistake is always the same: commission first, then want funding. That doesn't work. Allow several weeks of lead time for application and approval.

04 — Scope

What's included

• Inventory: which AI tools are actually being used in the business
• Classification under the EU AI Act — risk class and resulting obligations
• Review of the data protection situation for each tool, including data processing agreements
• Usage policy that's practical in everyday work rather than just formally correct
• Implement labeling requirements for chatbots and AI-generated content
• Proof of AI competence through documented training

06 — Error

The most common mistakes

• Starting too big. Projects where nobody can use anything for months fail more often than average. A small, usable first step is almost always the better way.
• Not involving the people who work with it. Requirements that only come from management rarely match the actual work.
• Not defining a success measure. Without a metric defined in advance, you can't say afterwards whether it delivered anything — and the next decision will be made on gut feeling again.
• Overlooking operating costs. The project price is only part of it. What operation costs over three years belongs in the same calculation.
• Creating dependency. If you don't get access, documentation and source code, you're locked in. That needs to be clarified before commissioning, not after.