Regional AI Forum in Latin America and the Caribbean Expands Reflection on Opportunities and Challenges of Artificial Intelligence

Organized by UNESCO and co-organized by CGI.br, NIC.br, and the Federal Government, the Forum gathered renowned authorities and experts at USP

Key issues associated with Artificial Intelligence (AI) such as governance, ethical dimensions, and personal data protection were discussed by national and international authorities and experts during the Regional AI Forum in Latin America and the Caribbean, held on Thursday (12) and Friday (13) at the University of São Paulo (USP). The UNESCO event was co-organized by the Brazilian Internet Steering Committee (CGI.br) and the Brazilian Network Information Center (NIC.br) through its Regional Center for Studies on the Development of the Information Society (Cetic.br), and by the Federal Government through the Ministry of Science, Technology, Innovations, and Communications (MCTIC) and the Ministry of Foreign Affairs (MRE), in partnership with USP.

During the opening ceremony, Vahan Agopyan (USP) praised the Forum's choice to bring a human approach to the AI discussion. Complementarily, Minister Marcos Pontes, from MCTIC, highlighted that "artificial intelligence is a very powerful tool, but it must be under human decision and command." Pontes emphasized the importance of discussions involving different countries and cultures to find solutions to common problems.

"It is essential to understand the reach of AI, possible changes in human behavior, the logic of algorithms, deepen surveys, and promote the construction of a national AI strategy," noted Maximiliano Martinhão (CGI.br), who welcomed UNESCO's initiative in holding meetings that facilitate defining the roles of countries in face of the challenges and opportunities of AI. Agreeing, Moez Chakchouk (UNESCO) pointed out that ethics in AI must be debated at an international level, including various stakeholders from developed and developing countries.

During the opening ceremony, there was also the signing of an agreement amendment between MCTIC, FAPESP, and CGI.br for implementing AI labs in the country. "The Internet in Brazil was born at FAPESP, which was the cradle of NIC.br. This is FAPESP's role: to promote initiatives that grow beyond it. I am also happy about the signing of this agreement to create research centers at the frontier of AI and cybersecurity knowledge," stated Marco Antonio Zago (FAPESP). The project involves implementing eight centers, half of them in São Paulo and the other half in other Brazilian cities, which should concentrate researchers from various areas.

AI and Public Policies

In an inter-American panel on developing public policies for AI, Minister Marcos Pontes highlighted the public consultation launched by MCTIC for civil society's participation in the Brazilian Artificial Intelligence Strategy. Pontes listed policies implemented by the Federal Government to form the basis for introducing new technologies into the country, such as the decree for the Internet of Things (IoT), which has four discussion chambers: industry, health, smart cities, and agriculture.

The panel included Jennifer May, Canada's ambassador to Brazil, who commented on the Canadian AI policy. Canada was the first country to develop a strategy in the area. "Support for innovation, commercialization, attention to implications, all based on collaboration. These are the main points that compose the Canadian AI strategy, in addition to the development of leading research centers," she stated. Chilean policies to connect companies, universities, and the government were addressed by José Antonio Guridi (Ministry of Science, Technology, Knowledge, and Innovation of Chile), while Diego Fernández Montero (Chief Data Officer of the Government of Costa Rica) emphasized the importance of creating a culture of data-based decision-making and public policy formulation.

To understand the strategies, frameworks, and principles already developed at the regional and international level regarding value-centered artificial intelligence, among other issues, one of the panels of the Forum specifically addressed international cooperation. Representatives from OECD, ITU, ECLAC, IEEE, and the UN High-Level Panel on Digital Cooperation presented initiatives and guidelines formulated by these international organizations. Valeria Jordan (ECLAC) highlighted the need to consider technological advances from more developed countries while also taking into account each country's particularities. "Citizens are an important part of this equation. It is essential to ensure they have a minimum level of understanding about artificial intelligence so they can engage in discussions and assert their rights," defended Sacha Alanoca (AI Civic Forum).

For Nicolas Miailhe (The Future Society), the recommendations of OECD, UNESCO, and IEEE should be celebrated. "They converge on five principles: transparency, fairness and equity, non-maleficence, responsibility, privacy. However, these recommendations have their limits. There is a lack of clarity in terminologies, geographic diversity, and civil engagement," he warned. Nicolas also called attention to the ethical challenges of AI, which involve not only privacy but also discrimination, information asymmetry, work automation, and limitation of freedom of expression and choice.

Work automation was also addressed by Cristina Gómez (Center of Excellence in AI of Medellin, Colombia), who discussed the incorporation of worker robots in business processes and human-robot interaction. "When we develop artificial intelligence systems that have the capacity to generate more empathy with others, we will be able to generate better coordination and cooperation between machines and humans," she opined.

AI and Transparency

The impacts of technologies such as AI, IoT, and big data on freedom of expression, access to information, and knowledge were also discussed during the Forum. Isabela Ferrari (Federal Justice) highlighted unprogrammed algorithms that use machine learning to discover expected results through a series of correlations. "This creates an enormous difficulty for humans to articulate the logic taken by this type of software. The machine learning algorithm will bring the learning rule of the algorithm; it's no use opening the code. This is a major challenge: accountability needs to accompany the systems from the moment they are born," she stated.

Complementarily, Sara Rendtorff-Smith (MIT) brought examples of issues arising from AI use, particularly regarding discrimination, such as a system used to detect unemployment insurance fraud in the state of Michigan (USA), where 85% of fraudulent cases identified were incorrect. "Humans have biases. Artificial intelligence had and still holds the promise of correcting this, but it is not happening. We are concerned that AI does not exacerbate these issues," she cautioned.

José Luiz Ribeiro (RNP and CGI.br) emphasized the importance of combating these problems. "Perhaps we need to develop, in analogy to antibodies, anti-algorithms, which will monitor, oversee, and tackle AI deviations at a time when our capacity is still very slow. It is necessary to discuss and create ways to discipline the decisions and undue signals of the AI models we are creating," he said.

During the event, Demi Getschko (NIC.br and CGI.br) stressed that AI regulation discussions should be principle-based. Still, in Getschko's evaluation, decisions should remain in human hands. "When you pass the decision over to the system, you change the level. The automated legal system, for example, helps in data access, but when a judicial decision becomes the system's, we are leaping into something dangerous. The system can assist as long as the decision and responsibility are human," he emphasized.

How to educate children to understand AI and be the future developers of this technology? Jasmina Byrne (UNICEF) argued that children should be consulted and included in the process of AI policy discussions. The expert also drew attention to the myth that all children are digital natives. "It is essential to invest in digital literacy, preparing children to understand what is done with data collection. Children, parents, and guardians are not aware of how this data is used to profile their consumption and promote targeted advertisements. AI policies must consider this issue," she addressed.

AI-based educational systems, the learning and necessary skills, as well as the future of scientific research and environmental management, were among the topics discussed in depth during the Regional AI Forum in Latin America and the Caribbean, which also addressed the potential of AI in relation to the Sustainable Development Goals (SDGs). The event was streamed live on the Internet and is available in full on NIC.br's YouTube channel, watch: https://www.youtube.com/playlist?list=PLQq8-9yVHyOa8vVdk3MpnVk8AJDyhD7Ia.

About Cetic.br
The Regional Center for Studies on the Development of the Information Society, part of NIC.br, is responsible for producing indicators and statistics on the availability and use of the Internet in Brazil, disseminating periodic analyses and information on the network's development in the country. Cetic.br is a Regional Center under the auspices of UNESCO. More information at https://www.cetic.br/.

About the Brazilian Network Information Center – NIC.br
The Brazilian Network Information Center — NIC.br (https://www.nic.br/) is a civil, private, non-profit entity that, in addition to implementing the decisions and projects of the Brazilian Internet Steering Committee, has among its functions: coordinating the domain name registration — Registro.br (https://www.registro.br/), studying, responding and handling security incidents in Brazil — CERT.br (https://www.cert.br/), studying and researching network and operations technologies — Ceptro.br (https://www.ceptro.br/), producing indicators on information and communication technologies — Cetic.br (https://www.cetic.br/), implementing and operating the Internet Exchange Points — IX.br (https://ix.br/), enabling the Brazilian community’s participation in the global development of the Web and supporting the formulation of public policies — Ceweb.br (https://www.ceweb.br), and hosting the W3C office in Brazil (https://www.w3c.br/). 

About the Brazilian Internet Steering Committee – CGI.br
The Brazilian Internet Steering Committee, responsible for setting strategic guidelines related to the use and development of the Internet in Brazil, coordinates and integrates all Internet service initiatives in the country, promoting technical quality, innovation, and the dissemination of offered services. Based on the principles of multisectorialism and transparency, CGI.br represents a democratic Internet governance model that is internationally praised, where all sectors of society participate equally in its decisions. One of its formulations is the 10 Principles for the Governance and Use of the Internet (https://www.cgi.br/principios). More information at https://www.cgi.br/.

Media Contacts:

Weber Shandwick
https://www.webershandwick.com.br/
PABX: +55 11 3027-0200 / 3531-4950
Marina Faria – mfaria@webershandwick.com - +55 11 3027-0241
Victor Sartori - vsartori@webershandwick.com - +55 11 3027-0223
Carolina Boccardo - cboccardo@webershandwick.com - +55 11 98708-0117

Communication Office – NIC.br
Caroline D’Avo – Communication Manager – caroline@nic.br
Carolina Carvalho – Communication Coordinator – carolcarvalho@nic.br
Soraia Marino – Communication Assistant – soraia@nic.br 

Flickr: https://www.flickr.com/NICbr/
Twitter: https://www.twitter.com/comuNICbr/
YouTube: https://www.youtube.com/nicbrvideos
Facebook: www.facebook.com/nic.br
Telegram: www.telegram.me/nicbr
LinkedIn: https://www.linkedin.com/company/nic-br/
Instagram: https://www.instagram.com/nicbr/

 

This content was automatically translated with the support of artificial intelligence.