The first proposal of the high-level group for artificial intelligence was published today and can be commented on by all 18 January. The final version should be presented in March.
Europe wants to make a difference in artificial intelligence for the capacity for development but also in ethics, and the various bodies have argued that it is important that any application of technology be sustained by common European values and fundamental rights defended by society.
Fears of misuse of Artificial Intelligence, whether in robotics or in privacy and security, should not detract from the benefits that technology can bring in the various areas and so the high-level expert group integrated in the European Alliance for AI, to define the ethical "script" to be applied in the development of new solutions, which can lead to an application framework.
The group, which includes 52 academics, business and civil society experts, sets out in these guidelines how IA should be applied in order to respect fundamental rights, applicable regulations and principles defended in the European Union, and how it can be guaranteed robustness and reliability of the technology.
The search for a cure for cancer or the reduction of energy consumption are some of the possible applications of Artificial Intelligence but the European Alliance for AI is coordinating the debate on the ethical component, considered of extreme importance and a differentiator of European solutions compared to what is done in the United States and China, which are already more advanced in the technological development and its application in the market.
In framing a reliable IA, Europe wants to ensure that there is a centralization in the human being, with fundamental rights, social values and the principle of benefit, not use to harm citizens. In this regard it is important to pay particular attention to more vulnerable groups such as children, minorities or asymmetries in access to power or information, but also to keep vigilance over the possibility of uses for evil purposes.
Responsibility, data governance, design for all, governance of human autonomy with human supervision, non-discrimination, respect for human autonomy, respect for privacy, robustness, security and transparency are the 10 key areas highlighted.
The guidelines for ethics in AI are open for consultation and comment until January 18, and the contributions will be collected and debated, to be included in the final document to be presented in March 2019, when it will be presented to the European Commission .