Author: Robayet Hossain ‘26
Editor: Andrew Ni ‘26
Source: McKinsey & Company
A project by the city of São Paulo, Brazil, known as SmartSampa, aimed to deploy 20,000 cameras with the ability to identify individuals through their “body structure” and monitor social media. However, the project faced significant backlash as a report from the city’s audit court stated that the project violated citizen rights, particularly those from the black and LGBTQIA+ communities. This case serves as a stark illustration of the concerning trend in which state-sponsored technology for public safety is increasingly susceptible to misuse and abuse by government administrations and councils.
Artificial intelligence (AI) is rapidly transforming the world, and South American countries are no exception. Governments, businesses, and academia across the region are investing heavily in AI research and development, and AI is already being used in a wide range of applications, from healthcare and education to agriculture and finance. However, with the rapid development of AI comes several ethical and moral challenges. For example, how can we ensure that AI systems are fair and unbiased? How can we prevent AI from being used for malicious purposes? And how can we guarantee that AI benefits all of society and not just a privileged few?
In South America, nations are taking the lead in addressing these issues and rectifying their previous history of technology misuse. The Brown University AIRES (AI Research, Ethics, and Society) Chapter has worked closely with the Pontifical Catholic University of Rio Grande do Sul (PUCRS) in Brazil with remote research focused on public databases on recent machine learning models and their potential risks, as well as mapping AI ethics across the globe. In recent years, several countries in the region, just like Brazil, are in the process of developing more responsible, ethical, and progressive applications of AI use for the betterment of their public societies [2]. Although the progress is exponential compared to the past, there is still a long way to go.
Brazil
In 2019, Brazil became the first country in the world to adopt a national AI ethics framework outlining a set of principles guiding the development and use of AI in the country, including principles of fairness, transparency, and accountability [2]. Then, in 2021, the government launched a public consultation on its draft AI ethics framework; this consultation received over 10,000 responses from individuals, organizations, and civil society groups, suggesting that there is a high level of public interest in AI ethics in Brazil [2].
Argentina
In Argentina, the government is using AI to develop a new system for predicting and preventing crime. The system is designed to identify individuals who are at risk of committing crimes and prevent them with early intervention services [3]. The system is still in its early stages of development, but it has the potential to revolutionize the way that crime is prevented in Argentina [3]. For example, the system could be used to identify individuals with connections to crime syndicates or past history who are at risk of committing domestic violence or gang violence. The government could then provide these individuals with early intervention services, such as counseling or job training, to help them pursue a better path [4].
However, there are concerns that the system could be used to target and discriminate against certain groups of people. Guillermo Ibarrola, whose story was documented in an article, spoke about an incident where such a facial recognition system, which critics call “the twisted eye in the sky of Buenos Aires” identified him as a criminal for an armed robbery [8]. The twist in this case? Ibarrola had never been in the city where the alleged crime took place; he was held six days in a large station hall on bare concrete and was told his possible sentence if convicted could reach 15 years [8]. The correct culprit of the crime—a different, slightly older Guillermo Ibarrola—was eventually discovered, but this sort of mistake could have been extremely costly and completely change the course of one’s life
Colombia
In Colombia, the government is using AI to develop a new system for personalizing education. The system is designed to identify each student’s strengths and weaknesses and create a personalized education plan accordingly [5]. The system is still in its early stages of development, but it has the potential to improve the quality of education for all students in Colombia. For example, the system could be used to identify students who are struggling in a particular subject and provide them with extra help [6]. Or, the system could be used to identify students who are excelling in a particular subject and provide them with more challenging coursework [6]. At the same time, a huge disadvantage of AI in the country’s education system is the existence of the mass collection and analysis of data on students, including their performance, behavior, and personal information, which poses a huge ethical and privacy concern [5].
Benefits and Challenges
AI ethics can help to ensure that emerging technology is used responsibly, builds public trust, and protects the rights and freedoms of individuals. Besides the examples mentioned above, researchers within Brazil have also developed a system capable of diagnosing dengue fever with 95% accuracy and Argentinian researchers now have a way of predicting crop yields with 85% accuracy [7,9]. While the potential benefits of AI ethics are clear, some challenges need to be addressed. For example, AI can often be complex and difficult to implement, could stifle innovation and creativity, and could justify discriminatory or unjust practices.
One challenge is that AI ethics is a complex and evolving field. There is no one-size-fits-all approach to AI ethics, and what works in one context may not work in another. An AI system that is used to predict and prevent crime could be biased against certain groups of people, such as the poor or the elderly. In the end, the potential impacts of AI ethics on South American countries highly depend on how AI ethics frameworks and policies are developed and implemented. If AI ethics frameworks are developed and implemented thoughtfully and inclusively, they can help ensure AI is used in a responsible and ethical manner that benefits all of society.
Conclusion
Imagine a world where AI is used to predict and prevent crimes before they happen. Or a world where AI is used to develop new medical treatments that are tailored to individual patients. Or a world where AI is used to create personalized education plans for each student. This is the future that South American countries are working towards, as they move towards the development and implementation of AI ethics frameworks and policies.
The potential benefits of AI ethics are clear. AI ethics can help to ensure that AI is used to solve real-world problems and improve the lives of people in South America [5]. However, it is important to address the challenges of AI ethics, such as the complexity of the field and the potential for misuse. South America is in the midst of this challenge as they develop AI ethics frameworks that will ensure this technology is used for good, to heal the sick, feed the hungry, and protect the planet. Their vision is a world where AI is not a threat, but a partner in our quest to build a better future—a world where AI helps us overcome our challenges and realize our full potential.
Works Cited
[1] The US blueprint for an AI bill of rights: Putting enough guardrails in place for Ai Ethics [Internet]. Commtel Networks; 2022 [cited 2023 Oct 30]. Available from: https://www.linkedin.com/pulse/us-blueprint-ai-bill-rights-putting-enough-guardrails-/?trk=organization_guest_main-feed-card_feed-article-content
[2] Ethics of Artificial Intelligence (AI) in Brazil [Internet]. Unesco.org. 2022. Available from: https://www.unesco.org/en/fieldoffice/brasilia/expertise/artificial-intelligence-brazil
[3] AI: Argentine Intelligence. A Conversation about Artificial Intelligence | Programa De Las Naciones Unidas Para El Desarrollo [Internet]. UNDP. [cited 2023 Nov 1]. Available from: https://www.undp.org/es/argentina/blog/ai-argentine-intelligence-conversation-about-artificial-intelligence
[4] AI Strategies and Policies in Argentina [Internet]. oecd.ai. Available from: https://oecd.ai/en/dashboards/countries/Argentina
[5] Teaching and Learning in the Age of AI: Considerations, Resources, and Opportunities [Internet]. ctl.columbia.edu. [cited 2023 Nov 1]. Available from: https://ctl.columbia.edu/resources-and-technology/resources/teaching-learning-ai/
[6] BKC Scholars Helped Guide Colombia’s AI Policy Implementation as Part of the Country’s “AI Expert Mission” | Berkman Klein Center [Internet]. cyber.harvard.edu. 2023 [cited 2023 Nov 1]. Available from: https://cyber.harvard.edu/story/2022-08/bkc-scholars-helped-guide-colombias-ai-policy-implementation-part-countrys-ai-expert
[7] Cabrera M, Leake J, Naranjo-Torres J, Valero N, Cabrera JC, Rodríguez-Morales AJ. Dengue Prediction in Latin America Using Machine Learning and the One Health Perspective: A Literature Review. Tropical Medicine and Infectious Disease [Internet]. 2022 Oct 21 [cited 2022 Dec 14];7(10):322. Available from: https://www.mdpi.com/2414-6366/7/10/322
[8] Naundorf K. The Twisted Eye in the Sky Over Buenos Aires [Internet]. Wired. [cited 2023 Nov 1]. Available from: https://www.wired.com/story/buenos-aires-facial-recognition-scandal/
[9] AE571/AE571: Artificial Intelligence (AI) For Crop Yield Forecasting [Internet]. edis.ifas.ufl.edu. Available from: https://edis.ifas.ufl.edu/publication/AE571
Comments