Big Data and Artificial Intelligence. An Approach from a Legal Point of View about Fundamental Rights

Abstract

Some dangers of big data and artificial intelligence are warned: data and non-objective algorithms can have important biases or "embedded" values. Guarantees against discrimination and "algorithmic bias" are affirmed, in the scope of privacy impact assessment or non-discrimination by default and by design. Some reflections are made on the massive personalization of information, the loss of the public sphere and the post-truth, as well as the freedom of expression of machines. Also a right to "algorithmic transparency" is intended to affirm and base. Different questions are exposed about the use of Big Data and artificial intelligence in law enforcement and justice. Finally, about privacy and data protection, it is stated that the guarantee of consent is outdated and ineffective and the difficulties of projecting the principles of data protection, as well as the importance and problems of anonymization are also affirmed.