By Toby Sterling

THE HAGUE The Netherlands and the U.N. on Thursday launched a project to help prepare Europe's national agencies to supervise artificial intelligence (AI).

The project, undertaken with support of the European Commission, comes ahead of the passage of the AI Act, the broad legislation that is expected to govern AI use in Europe.

In the project, The United Nations' Educational, Scientific and Cultural Organization (UNESCO) will be assembling information about how European countries are currently supervising AI and putting together a list of "best practices" recommendations.

The Dutch digital infrastructure agency (RDI) will be assisting UNESCO in communicating and meeting with national working groups from around Europe, including a first meeting on Thursday in The Hague.

"What we learn here, I promise you, is going to be informing many discussions that are happening at the very same time everywhere around the world," said UNESCO's Gabriela Ramos.

UNESCO has been an important voice in debates on AI, with ethical guidelines it put forward in 2021 adopted by all 193 member states, though they are not legally binding.

Nathalie Berger, director at the European Commission service that helps states enact reforms, said the AI Act will be passed this year, but it will take an additional two years to come into full force.

"Supervision is absolutely key to make sure that everything works well in practice," she said.

(This story has been corrected to fix Berger's title in paragraph 7)

(Reporting by Toby Sterling; Editing by Alexandra Hudson)