“From a legal perspective, this is a complicated area, as the algorithms of an AI act as a “black box” – even the creator of the artificial intelligence is not always sure why the algorithm made a certain decision. There are no classical cause-and-effect relations that we are used to seeing in existing software. We can only keep improving the algorithms, until their capability at a certain decision point is either as good or even better than that of a human,” Kaevats explained.
Kaevats added that the legal area of the implementation of systems without human interference is a challenge and an unresolved issue in technological development – the question is, who is responsible for the results of the decisions made by systems?
Siim Sikkut, Deputy Secretary General for Communications and State Information Systems of the Ministry of Economic Affairs and Communications, highlighted that the implementation of artificial intelligence could have various benefits for Estonia. “In the public sector, it would allow us to increase the user-centeredness of services, improve the process of data analysis, and make the country work more efficiently by achieving the goals of developing the e-government,” Sikkut emphasised. “Artificial intelligence can also play an important role in the digital revolution of the industry and attract new investments and innovation activity to Estonia – developers of technology are searching for a development and test environment that favours artificial intelligence solutions.”
The work of the expert group is coordinated by the Government Office in cooperation with the Ministry of Economic Affairs and Communications and its term is April 2019.
The expert group will comprise state authorities, universities, companies, and independent experts.
Estonian Government, March 2018