AI Act in IT: system changes and the company's role
We check which AI systems you build and use, what role the company has in each, and which change needs a new assessment.
AI Act and GDPR audit, paid once
from €700 net
Review and written summary: €0
Audit in 5 working days
How much does an AI Act audit for an IT team cost?
Short answer
An AI Act audit for an IT team costs from €700 net. It is ready in 5 working days. In our scenario, reviewing the AI systems yourself takes 130 h of work by IT, the data protection officer and a manager. That review costs €4,537. A lawyer assesses the company's role and obligations. Buying a ready-made model does not settle the role for every application.
Statistics
AI Act and IT teams: statistics 2026
2 August 2026
According to the European Commission (2026), the transparency obligations in art. 50 of the AI Act, including telling people they are talking to AI, apply from 2 August 2026.
According to the European Commission (2026), the rules for high-risk AI systems in Annex III of the AI Act, which include recruitment and education, apply from 2 December 2027.
one-off, to review the AI systems yourself: 130 h of work by IT, the data protection officer and a manager
On your own€4,537
With the auditfrom €700
Difference: up to €3,837
In your company
Team of 5, 3 systems
60 h of work · €2,094 one-off
from €700 with the audit
Department of 20, 6 systems
130 h of work · €4,537 one-off
from €700 with the audit
Software house, 12 systems
260 h of work · €9,074 one-off
from €700 with the audit
From 28 October 2026, compliance with the AI Act in Poland may be supervised by the Commission for the Development and Safety of Artificial Intelligence (KRiBSI). Under art. 99 of the AI Act, breaching obligations, including transparency, can cost a fine of up to €15 million or 3% of annual turnover, and for small and medium-sized companies the lower of the two.
On your own: inventory of tools and accounts 32 h, settings of each tool 32 h, processing agreements and data 22 h, map of AI Act obligations 26 h, AI policy and training plan 17 h. In total 130 h × €34.90 = €4,537. The audit: from €700 net in 5 working days. The difference: up to €3,837, plus the time you would spend learning the rules.
Assumptions: The hours and how they split for a company of this size are our assumptions. €34.90 an hour is the EU average hourly labour cost for 2025 according to Eurostat, for companies with 10 or more employees. The audit price grows with the number of tools, accounts and departments; for larger companies we quote it after the free review.
What stays with people: A lawyer assesses the company's role and obligations. Buying a ready-made model does not settle the role for every application.
foap.com customer service: 95% of replies are sent automatically.
Users from several countries write to foap.com, in several languages. The system reads the entire conversation and the terms of that person, and then sends a response in that person's language. Questions about money and exceptions go to the team with a ready draft, so they have to check and send them.
the system sends the answers itself, without human typing
95%
the system sends the answers itself, without human typing
time from the arrival of the e-mail to the ready response (median)
30 s
time from the arrival of the e-mail to the ready response (median)
e-mails that require a response receive a ready-made draft
100%
e-mails that require a response receive a ready-made draft
less work for every 1000 answers (estimate)
47 h
less work for every 1000 answers (estimate)
Data from foap.com production, October 20-21, 2026. Estimate: 3 minutes for handwritten response.
It organises a company's AI tools with owners, evidence and a list of gaps. A person approves the risk classification. On 14 test entries the scope matrix passed 9 of 9 checks. Demonstrator on test data.
“We started with my own AI agent. During the training that followed, I learned to build new applications myself. Today I am bringing this way of working with AI into my business. Working with Syntalith gave me exactly the skills and tools I needed.”
Register of AI tools with an owner and the scope of data
List of settings changes for the administrator
AI policy based on our template
Training plan for art. 4 of the AI Act
Map of AI Act obligations with deadlines
What happens to your data
The audit prepares the facts and the settings. Your data protection officer or lawyer approves the legal assessment.
Where personal data is involved, we sign a data processing agreement first.
Your lawyer provides the legal opinion and represents you before the authority. We hand them the full set of materials.
After the audit you keep the register of AI tools, the list of settings changes, the AI policy, the training plan and the list of questions for your data protection officer. Team training on art. 4 of the AI Act: from €1,150 net a day.
Questions
How much does the AI Act and GDPR audit cost?
From €700 net, paid once, within 5 working days of receiving access. The price depends on the number of tools, accounts and departments; we give the fixed price after the free review. Team training on art. 4 of the AI Act costs from €1,150 net a day.
If we build on a model API, are we a provider?
That depends on the system's purpose and on who places it on the market. The audit records the facts about each system, and your lawyer assesses the role.
When does it not pay off?
When the team uses one AI tool on company accounts and already has an AI policy. Then our free AI policy template and a settings review by your administrator are enough.
What do we need to prepare?
A list of systems with AI, their purposes, the model vendors and the person who approves releases.
What our OpenAI and Anthropic partnerships mean for you.
OpenAI makes ChatGPT and Anthropic makes Claude. We belong to the partnership programs of both companies, so we work with their tools every day and learn from the creators themselves.
Claude Partner Network
01We will select a tool for the task
We implement both ChatGPT and Claude. We check both on your documents, and you choose the one that makes fewer corrections and saves more time.
02Knowledge straight from the creators
We learn from official OpenAI and Anthropic courses and materials. Your team learns about the features and settings that work today.
03Licenses without our margin
You buy accounts directly from OpenAI or Anthropic and receive an invoice from them. We do not add anything to the license.
04Certified engineers
On the team, we have an AWS certification in machine learning and an Anthropic certification in systems architecture on Claude.
Participation in programs does not mean that OpenAI or Anthropic endorses our services.
customer service
