Trustworthy Artificial Intelligence and Human Rights

• Autores: Migle Laukyte
• Localización: Current Issues On Human Rights / coord. por Alexander Sungurov, Carlos R. Fernández Liesa, María del Carmen Barranco Avilés, María Cruz Llamazares Calzadilla, Óscar Pérez de la Fuente, 2020, ISBN 978-84-1324-552-2, págs. 69-80
• Idioma: inglés
• Texto completo no disponible (Saber más ...)
• Resumen

  • In April 2019, High-Level Expert Group on Artificial Intelligence, set up by the European Commission, has published its Ethics Guidelines for Trustworthy Artificial Intelligence (AI), which addresses the future of AI development in Europe. In particular, these Guidelines work out a vision of AI that Europe should foster and indicates the features that any AI-based system should have: the framework is composed of three parts, namely, lawful, ethical and robust AI. The Guidelines do not deal with the first of these parts—lawful AI—where, among other things, authors include the necessity that AI would respect fundamental human rights: this is the so called “fundamental-rights based approach” which is the approach that EU promotes with respect to AI.

    As the Expert Group prefers to focus on ethics and robustness of AI, this paper will focus on lawfulness of AI. In particular, I will focus on the impact that AI could have on human rights, established in the Charter of the Fundamental Rights of the European Union (EU), other relevant international treaties, and specific regulations such as General Data Protection Regulation. The Guidelines list some of the human rights that should be the foundational stones for any AI within the EU, namely, protection of human dignity and human freedoms in the broad sense of the term, respect for democracy, justice and the rule of law, equality and non discrimination and citizenship-related rights, such as the right to vote. These rights also reverberate in the discussion on ethical AI, where specific ethical principles are being discussed.

    The aim of this paper is to address the aforementioned human rights and see in what ways AI could have an impact on them: how could AI not only respect (passive stance) but also support and help to bring into being (active stance) some of these rights? In particular I will look at the existing AI applications and discuss whether we are approaching the challenge to make AI work for (and not against) human rights in the right way. Furthermore, the paper will also raise the question whether AI could advance any kind of new human rights that we might consider to be fundamental in the future. For instance, do we have a right to know when we interact with an AI on telecommunication networks and not with a human being? Do we have a right to explicability of algorithms?