Twenty-six articles. A perimeter of rules that defines the rules of the game that Italy will play on artificial intelligence. The bill approved yesterday by the Council of Ministers contains many confirmations and few changes compared to the draft circulated in recent weeks. There are few restrictive interventions. More substantial are those to favor an AI industry which in Italy largely needs to be built. To facilitate this process, said the undersecretary to the Presidency of the Council, Alessio Butti, “an investment of around one billion is expected with the collaboration of Cpd”. Money which, said the Minister of Business and Made in Italy, Adolfo Urso, will be used to create and grow “startups operating in the sector and allow the birth of a national champion in the sector”.
And it is in the direction of facilitating the birth of an AI supply chain that the decision to give the Agency for Digital Italy (Agid) and the Agency for National Cybersecurity (Acn) the role of driving force respectively and controller of AI development in Italy. The choice of two government agencies has created some discontent. But the text, said Butti, “is absolutely in line with what was voted by the European Parliament, the Italian one is the first government to legislate on artificial intelligence”.
A bill which according to Butti “unequivocally defines who develops the strategy” and “who monitors, who supervises and who notifies and sanctions”. The bill starts from some general principles. Such as that of developing artificial intelligence that creates tools “respecting the autonomy and decision-making power of man”. Or, again, that they do not jeopardize “the democratic conduct of institutional and political life”. Principles expressed on work, on justice, on publishing where AI can be applied “without prejudice to the freedom and pluralism of the media, to freedom of expression, to the objectivity and completeness, impartiality and loyalty of information” .
No access to these technologies for children under 14 years of age, without parental consent. While anyone who causes harm to people by disseminating modified videos or audio will be punished with “imprisonment from one to five years”, recalled the Minister of Justice, Carlo Nordio. The Minister of the Seal reiterated the need for the provision, because “we find ourselves faced with a real revolution whose outcome we do not know”.
Perhaps the bravest part of the bill concerns healthcare. The processing of data, including personal data, carried out by public and private non-profit entities, is declared “necessary for the purposes of the creation and use of databases and basic models”, therefore, “of significant public interest”. Personal data can therefore be used to train models capable of making diagnoses and imagining treatments, “without prejudice to the inspection and sanctioning powers of the Authority for the protection of personal data”.
The reason is to allow research and enhancement of these tools. And again to strengthen skills in the AI field, the bill extends the tax relief regime to allow researchers who have carried out research activities in the sector abroad to return to Italy. Still on the training side, a curiosity: the text provides that students with “high cognitive potential” can follow higher training courses and be recognized with training credits. Bringing to Italy a practice already adopted for some time in the USA for the intellectual development of new talents.
#billion #startups #prison #harm #deepfakes #law
– 2024-04-24 15:30:28
