Artificial intelligence17 min read3238 wordsRedakcja dvnt
AI in business — where to start, what it costs and what the law says
AI in business without the hype: how many Polish firms use it, when an assistant is enough, what it costs and what the AI Act and Polish act require.

AI in business today means four different things: an assistant in your office suite, automation with a language model inside, a chat on your website and an agent that chooses its own next steps. Each has a different cost, a different risk and different legal duties, and under one name it is easy to buy something other than what the company needs.
In Poland fewer than one company in ten uses AI. According to GUS (Statistics Poland), in 2025 that was 8.7% of enterprises with at least 10 employees. Eurostat (11 December 2025) puts Poland at 8.4% against an EU average of 20.0%, which ranks it 26th of 27 countries. EU companies that considered AI but do not use it most often name lack of expertise, unclear legal consequences and data protection.
This guide helps you make three decisions: which kind of AI fits your problem, what it really costs, including outside the invoice, and which rules apply after the July 2026 amendment of the AI Act and the Polish act on artificial intelligence systems. Where a topic has its own article, we link to it. We checked the data, prices and rules on 3 October 2026; quotes from Polish sources are our own translation.
How many companies in Poland use AI — and for what
AI in business in Poland is still mostly a large-company affair: 42.0% of large companies use it, against 6.1% of small ones. That is GUS data for 2025 (enterprises with 10 or more employees, excluding the financial sector; small means 10–49 people, medium-sized 50–249, large 250 or more). Among medium-sized companies the share is 15.6%, and across all sizes 8.7%. A year earlier it was 5.9%, so the share is growing, but from a low base.
Most of it is about text. According to GUS, 5.3% of enterprises used technologies that generate written or spoken language, and the most common purpose was marketing and sales (5.0%), ahead of administrative work in HR and management (3.7%). The GUS publication "Społeczeństwo informacyjne w Polsce w 2025 r." (Information society in Poland in 2025, 16 December 2025) adds that the most common way of acquiring AI was "buying a commercial, ready-to-use solution (6.4%)". Our reading of these numbers: most companies that use AI buy a ready-made tool rather than build their own system, and more than half use it for text. The same publication reports that 91.3% of enterprises did not use AI and 3.9% considered using it.
Against the EU, the gap is widest among small companies. According to Eurostat (2025 data, read 3 October 2026), 17% of small enterprises in the EU used AI, against 5.79% in Poland. Among large companies the distance is smaller: 55.03% in the EU against 45.81% in Poland. Both institutions count companies without the financial sector (section K), and yet they give slightly different results for Poland: Eurostat 8.36%, GUS 8.7%. That is why we do not combine them in one calculation.
We found no data on AI agents in Polish companies. GUS and Eurostat measure "artificial intelligence technologies" in general, and surveys about agents, such as Gartner's or McKinsey's, are global and show no results for Poland.
Four kinds of AI in a business
One question decides which kind of AI you are looking at: who chooses the next step — a person, a pre-written workflow or the model. The answer determines the cost, the risk and the amount of work before launch.
Four ways to use AI in a business · who chooses the next step, what you pay for, the main risk
| Assistant in an office suite | Automation with a model inside | Chat on your website | Agent | |
|---|---|---|---|---|
| Who chooses the next step | Assistant in an office suiteA person chooses the next step | Automation with a model inside
| Chat on your website
| AgentThe model chooses the next step and the tool |
| What you pay for | Assistant in an office suitePer-user subscription | Automation with a model insideImplementation, automation tool and tokens | Chat on your websiteImplementation and tokens | AgentImplementation and tokens that grow with the number of steps |
| Main risk | Assistant in an office suiteThe assistant sees everything the employee can access | Automation with a model insideA wrong model output moves on unchecked | Chat on your website
| Agent
|
Assistant in an office suite. ChatGPT on a business plan, Copilot in Microsoft 365 or Gemini in Google Workspace help you write, summarise and search, but an employee decides every step. Gartner describes assistants (26 August 2025) as tools that "depend on human input and do not operate independently", and calls referring to them as agents the most common misconception. You pay a subscription per user. What the Google and Microsoft plans include, and what Copilot costs as an add-on, we compare in Google Workspace vs Microsoft 365.
Automation with a model inside. The workflow is written in advance and the model performs single steps in it, for example reading the content of a document. Anthropic, the maker of the Claude models, calls this a workflow: the model and tools are orchestrated "through predefined code paths" (Building effective agents, 19 December 2024). How automation differs from RPA and integration is covered in our article on business process automation, and the tools in which you can assemble such a workflow without a programmer in the piece on low code and no code. One such tool, in which the model is a step of the workflow, has its own article: n8n.
Chat on your website. It answers customers from the company's knowledge and hands the conversation to a person when it does not know the answer. It usually works as RAG (retrieval-augmented generation): before answering, it searches the company's documents for matching passages. How this works and what spoils such answers is covered in our article on RAG. It has two risks an internal assistant does not: the model can make up an answer that the customer takes for fact, and from 2 August 2026 the person on the other end must know they are writing to an AI (Art. 50 of the AI Act, explained below). How Polish shoppers use chatbots, and where a chatbot helps, is covered in our article on ecommerce customer service.
Agent. The model chooses the next steps and tools itself. Anthropic writes that agents are systems where models "dynamically direct their own processes and tool usage". OpenAI, in its guide to building agents (2025), states that a simple chatbot is not an agent. Gartner estimates (25 June 2025) that of the thousands of "agentic AI" vendors only about 130 are real, and that many rebrand existing products such as assistants, RPA and chatbots. Anthropic warns that an agent's autonomy means higher costs and the potential for compounding errors. What separates an agent from a chatbot and from ordinary automation, and when it really makes sense, we explain in the article on the AI agent in business, and how to build a first version — from one process to approvals, limits and tests — in the guide on how to build an AI agent.
Gartner gives a simple rule of thumb in the same press release: agents where decisions are needed, automation for routine workflows, assistants for simple retrieval.
Where to start — the order that saves money
Start with what blocks most companies: knowledge, rules and data, and choose the tool last. The barriers show in both surveys. GUS asks about the reasons of companies that considered AI but did not use it (3.9% of all), and reports the results as a share of all enterprises: lack of staff or expertise 2.4%, costs too high 2.1%, lack of clarity about the legal consequences 2.1%, privacy concerns 2.1%, access to data or its quality 1.3%. Eurostat reports shares among EU companies that considered AI: lack of expertise was named by 70.89%, unclear legal consequences by 52.52% and data protection by 48.83%. The two sets of figures cannot be compared directly, but in both lack of expertise comes first, not cost.
That gives four steps.
- Rules for using AI. Decide which data may be typed into which tool and from which account. The Polish data protection authority, UODO (Urząd Ochrony Danych Osobowych), points out (6 February 2025) that generative AI processes huge amounts of data, which "may be used for purposes contrary to the user's original wishes, e.g. to further train the model or for marketing purposes" (our translation). How to list where company data lives and who can reach it is covered in company data security.
- The assistant in the suite you already have or can add. It is the cheapest way for the team to learn to work with a model. Before switching it on, tidy file permissions (below, in the section on costs).
- One process to automate. Pick a process that repeats and has a clear, checkable result. Anthropic advises "finding the simplest solution possible", which may mean not building agentic systems at all, and adds that for many applications a single model call with retrieval and in-context examples "is usually enough".
- An agent only when the workflow does not fit. OpenAI lists three conditions: complex decisions with exceptions, rules that are hard to maintain, and heavy reliance on unstructured data. If your process meets none of them, then in OpenAI's words "a deterministic solution may suffice", that is, ordinary automation. Once you do build an agent, the same guide says actions that are sensitive, irreversible or high-stakes should trigger human oversight until confidence in the agent grows. OpenAI's examples: cancelling an order, authorising a large refund, making a payment.
A test before step four: can you write down the rule an employee uses today to make the decision? If so, you need automation, not an agent. In our view the rule OpenAI states for agents holds at every stage: start small, validate with real users and grow capabilities over time.
What AI costs a business — three kinds of cost and one hidden one
The cost of AI in a business is a subscription, token charges and implementation work, and before all of them comes tidying up your data, which appears on no price list. Each grows by a different rule.
Subscription per user. Assistants in office suites cost a fixed amount per person per month. The bill grows with headcount, not with how heavily the tool is used. Current Google and Microsoft plans with AI are compared in Google Workspace vs Microsoft 365.
Tokens. When a model works in an automation, a chat or an agent, you pay for the amount of text processed, counted in tokens (pieces of words). That is how Anthropic and OpenAI bill. What tokens and a model's context window are, we explain in the article on large language models (LLMs). We calculated an example ourselves on the Claude Sonnet 5.5 price list ($2 per million input tokens, $10 per million output tokens, read 3 October 2026). Anthropic publishes prices in US dollars, so we leave them in dollars. One chat reply, with 3,000 input and 300 output tokens, costs $0.009. An agent task with ten model calls, averaging 8,000 input and 500 output tokens per step, costs $0.21, about 23 times more. A thousand such tasks a month come to $210. These are assumptions of the example, not a market average, and they ignore discounts for caching and batch processing. Gartner estimates (17 August 2026) that routing a task to an agentic reasoning model raises the model provider's inference cost at least fivefold compared with a basic chatbot interaction. The full token calculation for an agent, with a chart comparing a chat reply and a multi-step task, is in the article on the cost of an AI agent.
Implementation and maintenance. Integration with your systems, preparing a knowledge base, testing and later model changes are work whose price depends on scope. We found no independent study of AI implementation prices in Poland, so we quote no market figure.
The hidden cost: data and permissions. Buying a licence does not finish the preparation. Microsoft's Copilot documentation (read 3 October 2026) says the assistant only shows organisational data that individual users have at least view permission for, and tells you to make sure the right people have access to the right content. The same page states that prompts, responses and data accessed through Microsoft Graph are not used to train the foundation models. It cuts both ways (our example): if personnel files sit in a folder shared with the whole company, the assistant will find them for anyone who asks. That is why the first step in Microsoft's deployment guide (read 3 October 2026) is to fix oversharing: find sites and files shared too broadly, ownerless, unused or holding sensitive data, and then correct access. Gartner writes (11 May 2026) that without a clear understanding of the relationships and rules in an organisation's data, AI agents "cannot operate accurately and are far more likely to hallucinate". Those hours of tidying are an implementation cost even if nobody invoices them. The same principle applies to an AI chat built on company documents: permissions must be enforced by the retrieval itself, as we explain in the article on RAG. It is also true of an assistant or agent connected to company systems through MCP (Model Context Protocol): it sees and does whatever the account it uses allows.
Money also limits those who have already deployed AI. In McKinsey's global survey (25 August 2026, 1,719 respondents from 97 countries, self-reported) about 20% said AI operating costs, including tokens, constrained their use of AI, and 37% attribute at least some EBIT impact to AI. A third of respondents (36%) work for organisations with revenue above $1 billion, and the survey says nothing about small companies in Poland.
The law: the AI Act, the Polish act and GDPR in one calendar
The AI Act applies in stages, and the July 2026 amendment pushed the heaviest duties to December 2027 — but the duty to tell people they are talking to an AI has applied since 2 August 2026. The dates come from Regulation 2024/1689, the amending Regulation 2026/1744 (in force since 27 July 2026) and the Polish act on artificial intelligence systems (ustawa o systemach sztucznej inteligencji, Dz.U. 2026 item 1003). Duties by company role, fines for SMEs and a checklist are in the article EU AI Act for business.
The key dates in order:
- 1.08.2024 — the AI Act enters into force.
- 2.02.2025 — Chapters I and II apply, including Art. 4 on AI literacy and the prohibited practices.
- 2.08.2025 — the rules on general-purpose AI models and the chapter on penalties.
- 27.07.2026 — the amending Regulation 2026/1744 enters into force.
- 2.08.2026 — general date of application, including Art. 50 on transparency. According to the European Commission, from that day the AI Office and the authorities of the member states are responsible for implementing, supervising and enforcing the rules.
- 11.08.2026 — the Polish act enters into force. It makes the Commission for the Development and Security of Artificial Intelligence (Komisja Rozwoju i Bezpieczeństwa Sztucznej Inteligencji, KRiBSI) the market surveillance authority.
- 28.10.2026 — the chapters of the act on inspections, proceedings and penalties start to apply.
- 2.12.2026 — providers of content-generating systems placed on the market before 2.08.2026 must mark their output (Art. 50(2)). Two new prohibitions also start that day, on generating intimate material without consent and on material depicting the sexual abuse of children.
- 2.12.2027 — obligations for high-risk systems in Annex III, including AI in recruitment.
- 2.08.2028 — obligations for high-risk systems in Annex I (products covered by other EU legislation).
Chat and agent talking to people. Art. 50(1) obliges the provider of a system to make sure the person on the other end knows they are interacting with an AI, unless that is obvious. The information has to appear at the latest at the first interaction (Art. 50(5)). A breach can bring a fine of up to EUR 15 million or up to 3% of total worldwide annual turnover, whichever is higher, and for SMEs under Art. 99(6) the lower of the two. Whether a company that orders a chat from a contractor and runs it under its own brand is the provider in this arrangement depends on the case; if in doubt, ask a lawyer.
Training your staff. Art. 4 in the wording given by Regulation 2026/1744 says that providers and deployers of AI systems "shall take measures to support the development of AI literacy" of their staff, and that the obligation "does not require providers or deployers to guarantee any specific level of AI literacy of any individual". In its questions and answers on AI literacy the Commission adds that Article 4 does not entail an obligation to measure the AI knowledge of employees, and that from 2 December 2027 those who deploy high-risk systems from Annex III will have to put people with the necessary competence and training in charge of oversight.
Is your AI an "AI system" at all. In its guidelines on the definition of an AI system the Commission writes that it is not possible to determine automatically which systems fall within the definition, and that only some AI systems are subject to regulatory obligations. High risk depends on the use, not on the technology.
GDPR. The European Data Protection Board says in its Opinion 28/2024 (17 December 2024) that supervisory authorities should take into account whether the controller deploying a model carried out an appropriate assessment that the model was not developed by unlawfully processing personal data. A company deploying a model should therefore check this. In its ChatGPT taskforce report (23 May 2024) the Board notes that, in any case, "the principle of data accuracy must be complied with", even though a model can produce made-up output. In Poland, data protection is supervised by UODO, whose warning about data typed into generative AI is quoted above.
Articles in this section
Below are the articles in this section, followed by articles from other sections that cover AI as part of their own topic.
- AI agent in business: how an agent differs from a chatbot and from automation, the three conditions that justify one, token cost and the AI Act.
- EU AI Act for business: who is a provider and who a deployer, deadlines to 2028, the Polish act and KRiBSI, fines for SMEs and a checklist.
- Business process automation: how integration, RPA and automation with AI differ, GUS data on systems in Polish companies, KSeF and examples of automation in a small company.
- Low code and no code: when a platform without programming is enough, which limits you cannot see at the start and what you take with you if you want to leave.
- Google Workspace vs Microsoft 365: plans and prices of both suites, including Gemini in Workspace and Copilot Chat and paid Copilot in Microsoft 365.
- Ecommerce customer service: where a shop chatbot helps according to the Gemius 2026 survey of Polish shoppers, where it does not, and the duty to tell customers they are talking to an AI.
- Ecommerce automation: what to automate first, what Zapier, Make, n8n and Base cost, and how to calculate the return on automation in hours worked, not in promises.
How we work with AI at dvnt
We use OpenAI and Anthropic models through their APIs and advise starting with the simplest solution that solves the problem. We build automations in n8n and Node.js on OVH in the EU — we pick the location that fits the project. The scope of that service is described on our process automation page.
We are preparing an AI chat for our own website that answers from company knowledge and hands the conversation to a person. The first message tells the visitor that an AI is answering. A person takes over, among other cases, when the visitor asks for it, when the bot does not know the answer, for a complaint and for legal or financial topics, and when the monthly token budget is exceeded. The CRM receives a summary, not the full transcript. Where the model itself processes the data depends on the provider and platform, and we have not yet decided that. The chat is not live, so we publish no results. After launch we will watch it for at least four weeks before proposing a similar solution to clients.
Frequently asked questions about AI in business
Does a small business need AI?
Not necessarily. According to GUS, 6.1% of small companies in Poland (10–49 people) used AI in 2025. Across all companies, buying a ready-made solution was the most common route, and text-generating technologies were the most popular. Anthropic, the maker of the Claude models, advises starting with the simplest solution and writes that this sometimes means not building agentic systems at all. If your business has repetitive work with text, start with an assistant in your office suite and rules for using data; if you have a process with a clear rule, ordinary automation is enough.
Can I paste customer data into ChatGPT?
Not from a private account and not without rules set in the company. UODO, the Polish data protection authority, points out that data typed into generative AI tools may be used to further train the model or for marketing purposes. The European Data Protection Board notes that a model built on unlawfully processed personal data can undermine the lawfulness of its deployment. Check the terms of the business plan you use; Microsoft, for example, states in its Copilot documentation that prompts, responses and company data are not used to train foundation models. Write down which data may go into which tool from which account.
Since when does the AI Act apply?
The regulation entered into force on 1 August 2024 and applies in stages. The prohibited practices and Art. 4 apply from 2 February 2025, the rules on general-purpose models and penalties from 2 August 2025, and most other provisions, including the duty to tell people they are talking to an AI, from 2 August 2026. After the amendment by Regulation 2026/1744, the obligations for high-risk systems in Annex III apply from 2 December 2027 and those in Annex I from 2 August 2028. The Polish act on artificial intelligence systems has applied since 11 August 2026, and its provisions on inspections and penalties since 28 October 2026.
Do we have to train staff on AI?
Art. 4 of the AI Act, in its July 2026 wording, requires companies that use AI to take measures supporting their staff's AI literacy, but not to reach a specific level. The European Commission explains that the article does not oblige you to measure employees' knowledge. A separate duty concerns high-risk systems: from 2 December 2027 oversight of them must be entrusted to trained people.
How is AI different from automation?
By who decides the next step. In automation a workflow written in advance decides, even if a model performs single steps in it. When the model itself chooses the next steps and tools, it is an agent: you gain flexibility but pay with higher cost and the risk of compounding errors. Gartner advises using agents where decisions are needed, automation for routine work and assistants for simple retrieval.
Related articles
Artificial intelligence
AI agent in business — what it is and when it makes sense
An AI agent is a language model that chooses its own steps and tools. When it makes sense in a company, what it costs and what the AI Act and Polish law say.
Artificial intelligence
How to build an AI agent for your business — from process to first working version
How to build an AI agent for your company step by step: process, tools and permissions, human in the loop, cost limits, evaluation and prompt injection.
Automation and integrations
What is n8n — and when does it make sense for a business
What is n8n? An automation tool with AI nodes, in the cloud or on your own server. Licence, prices with a PLN guide, GDPR, and n8n vs Make and Zapier.
Artificial intelligence
Model Context Protocol (MCP) — what it is and what it means for a business
Model Context Protocol (MCP) connects AI assistants and agents to company systems. How an MCP server works, its risks, GDPR and when to build your own.
Artificial intelligence
Prompt engineering — what it is and how to write prompts at work
Prompt engineering at work: how to write prompts, templates for business processes, and when a Polish company needs RAG or an AI agent instead.
Automation and integrations
API — what is it? REST API, webhooks and OpenAPI explained for business
API explained with NBP, KSeF and VIES as examples: REST API, webhooks, OpenAPI, API keys and integration security for business.