The Ministry of Digital Affairs has published a new version of the draft Act on Artificial Intelligence Systems. The document introduces a number of simplifications for businesses, including solutions supporting innovation and more flexible rules for overseeing the AI market in Poland.
The updated draft of the Act on Artificial Intelligence Systems incorporates comments from public consultations submitted on the previous version. The new document will now be subject to renewed consultations, inter-ministerial consultations, and review.
From the text below you will learn, among other things:
- What is a regulatory sandbox?
- What will the new control rules be?
- What role should the Artificial Intelligence Development and Security Commission play?
Read also: AI Act comes into force on August 2, 2025 – what should be done to safely process and monitor AI in an organization
Regulatory sandbox - an opportunity for the SME sector
One of the key provisions of the new law is the establishment of a so-called regulatory sandbox. This is a special, safe space for testing modern AI technologies before their full market implementation.
- Free access. Micro, small, and medium-sized businesses will be able to use the sandbox free of charge.
- Financial support. The minister responsible for digital affairs will be able to finance scientific research and development work on AI from the state budget and EU funds.
- Open profile. The scope of technology sharing is to remain open to innovators, which is intended to foster the development of the Polish artificial intelligence ecosystem.
See also: 2026 – How AI, ESG and energy will reshape the European Market?
Remote inspections and lighter sanctions for companies>/h2>
The Artificial Intelligence Development and Security Commission will oversee the market.
- Remote supervision. By default, inspections are to be conducted remotely, limiting interference with ongoing company operations.
- Mitigation of penalties. Introduction of an agreement mechanism for extraordinary leniency. Businesses that quickly correct violations will be able to count on a reduced penalty.
- Individual opinions. Companies will be able to request individual opinions (formerly "interpretations"), which will increase legal certainty when implementing AI models.
Read also: Artificial intelligence and personal data - how to avoid falling into the regulatory trap of the AI Act and GDPR?
The Artificial Intelligence Development and Security Commission - a new, independent supervisory body
The Commission will be an independent body, technically overseen by the Ministry of Digital Affairs. The Commission's chairman will be appointed by the Sejm, with the Senate's consent, for a five-year term, guaranteeing the office's impartiality.
The body will include representatives of key institutions, such as:
- The President of the Office of Competition and Consumer Protection and the Polish Financial Supervision Authority (KNF).
- National Labour Inspectorate (PIP) and the President of the Office of Electronic Communications (UKE).
- Experts in the field of medicine and law (including the Office for Registration of Medicinal Products).
Security and transparency of new regulations
The new regulations place a strong emphasis on transparency. Parties filing complaints about AI systems will be informed of every stage of the proceedings. At the same time, limiting the number of ministerial representatives in the permanent composition of the Commission aims to strengthen its independence from current politics.
The draft law is currently undergoing further inter-ministerial and public consultations. By introducing these provisions, Poland is becoming a leader in the EU in streamlining regulations related to artificial intelligence.
Prepare your business for change. Find out how we can help.
Contact us.