Italian AI National Strategy: the plan drafted by the MISE’s Group of AI Experts

The Ministry of economic development (MISE) just published the National Strategy on artificial intelligence (AI). The document was drafted by a group of 30 experts in the field of artificial intelligence, in 1,5 year of work.

MISE considers the definition of an AI National Strategy a crucial priority for Italy, especially after the outbreak of Covid-19 and the subsequent crisis.

The strategy, which puts individuals at the centre and follows an approach that integrates technology and sustainable development, will (hopefully) help to boost public and private investments in AI and related technologies.

The document, with 82 recommendations, will be the basis for the definition of the Italian strategy within the European Coordinated Plan on Artificial Intelligence. Hence, it was prepared with the aim of addressing policies and tools in order to:

  • improve, coordinate and strengthen the research in the AI field;
  • promoting public and private investments in AI, also benefitting from the dedicated EU funds;
  • attract talents and develop business in the field of AI;
  • encourage the development of the data-economy, paying particular attention to the spreading and valorisation of non-personal data, adopting the better standards of interoperability and cybersecurity;
  • outline a comprehensive review of the legal framework with specific regard to safety and responsibility related to AI-based products and services;
  • conduct analysis and evaluations of the socio-economic impact of development and widespread adoption of AI-based systems, along with proposals for tools to mitigate the encountered issues.

The document is structured in three parts:

  • the first is dedicated to the analysis of the global, European and national market of artificial intelligence;
  • the second part describes the fundamental elements of the strategy;
  • the third part deepens the governance proposed for the Italian AI and proposes some recommendations for the implementation, monitoring and communication of the National Strategy on AI.

The group of experts estimated a public budget of € 888 million for the first 5 years to get the Italian AI strategy off the ground. Other € 605 million (€ 121 million per year) of private contributions are also needed.

The fields identified by the group of experts to invest in are: IoT, manufacturing, robotics, health care and financial services, mobility and transport, agrifood, energy, aerospace and defense, public administration, culture and digital humanities.

The plan also considers high-performance computing (HPC). With worldwide excellences such as Eni’s Green Data Center in Ferrera Erbognone and Cineca’s Leonardo supercomputer, Italy weighs 1.2% in the global HPC  panorama, with around 50 petaflops (PFlop). To double this calculation capacity, the plan proposes to invest € 70 million to reach 104 PFlop in 5 years.

Moreover, about half of the total budget, € 420 million, is dedicated to support 1,600 projects among PhDs, researchers and professors (40% co-financed by private institutions).

Experts suggest to create a group of specialists, reporting directly to the Italian Government, to execute the plan. Also, the creation of an Italian Institute for artificial intelligence (I3a) is suggested. The Institute can help to guide local studies and developments, attract international talents, and promote technology transfers between universities and companies.

Once implemented, the strategy needs € 80 million of annual funding, and about 1,000 people: 600 in the headquarters and 400 distributed throughout 35 research centres associated to Italian universities and research laboratories.

The strategy comprises other measures. For example: extension of sandbox to AI projects; compulsory insurance to cover AI responsibilities; creation of an out-of-court mechanisms to settle disputes with users who suffer from disservices; definition of a tax credit on research; etc.

The group of experts also suggest to merge various databases of the public administrations, and to create a unique regulated big data market (useful for training AI) and a central exchange platform. Finally, training at all levels (in schools and companies) is needed, as well as the establishment of national competitions to attract ideas and talents.


SUBSCRIBE TO THE SCALEUPITALY NEWSLETTER

The AI National Strategy is available (in Italian) at the following link: https://www.mise.gov.it/images/stories/documenti/Proposte_per_una_Strategia_italiana_AI.pdf