In January, the European Parliament's internal committee approved the new rules to solve challenges on the adoption and development of artificial intelligence. These rules aimed at a coordinated approach between Member States for the protection of citizens' data. The proposals were presented in plenary session on Wednesday and approved by the Member States.
The technology based on automatic decisions (ADM) is welcomed by the European Union, in order to innovate and improve services to consumers, especially at the level of new digital services, such as virtual assistants and chatbots. However, consumers must be informed when they are interacting with ADM systems and how they work, as well as providing access to human operators with decision-making capacity. Consumers have the right to ask for checks and corrections to automatic system decisions, managed by AI.
Companies must have structures that can be quickly activated to correct errors automatically decided. And again, maintaining human sovereign decision in the face of options taken automatically. Humans must always have the ultimate responsibility and ability to change decisions imposed by ADM systems, especially in medical, legal, banking and other matters.
On the other hand, Member States ask the European Commission for an update to the rules related to toys and mechanical products. Consumers must be informed about how to use these products and protect them from damage, forcing manufacturers to be clear in their obligations. The Product Responsibility Directive that was adopted 30 years ago should also be revised, in order to adapt concepts related to product defects and failures.
There are also plans for rules that prove the absence of information to the consumer, about the modification of the price of goods and services that are personalized through ADM systems. And that they do not do unjustified geo-blocking, which discriminate against consumers based on their nationality, residence or temporary locations.
The European Commission will present the final plans on the Artificial Intelligence approach next February 19th.