Challenge statement
Challenge type: If you are working on multiple challenges, please indicate if this is your "big bet" or "exploratory" challenge.
Please note: we ask you to only submit a maximum of 3 challenges - 1x Big Bet, 2x Exploratory. Each challenge must be submitted individually.
EXPLORATORY
Challenge statement: What is your challenge? (Please answer in specific terms: "Our challenge is that...”.)
In Argentina, our challenge is to promote digitalization, broaden the data community, foster collaborative data exploration and contribute to the public conversation on these developments, particularly the use and regulation of Artificial Intelligence (AI). This initiative will foster awareness of the potential of data and AI to accelerate progress towards the Sustainable Development Goals agenda. We are currently working on projects that promote digital literacy, particularly the ability to interpret data visualizations. At the same time, we aim to enhance understanding of the opportunities and risks associated with harnessing AI for sustainable development. Our long-term vision is to leverage collective intelligence to discover patterns within the vast ocean of data that surrounds us, generate evidence-based information, and identify future pathways for AI. Ultimately aiming for regulations that ensure a balanced approach, maximizing AI’s potential while mitigating its risks.
To address this challenge, the Lab plans a portfolio of actions. Knowledge products are currently being developed to explore the impact of AI on development, including its potential benefits, risks and emerging trends. Additionally, these resources will delve into the current state of AI-related regulations in Argentina. Other actions include organizing seminars, fostering AI-related capabilities within our Country Office, conducting prospective exercises, and supporting a data visualization contest called "Contar con Datos" (which translates to "Tell Stories with Data" and "Count on Data to Make Decisions").
Background: What is the history of your challenge? What is causing or driving it? Who is involved? How does the current situation look like? What undesired effects does it produce?
At the country level, policies and plans aimed at addressing the developments and challenges of Data and AI have been scarce. Since the new president took office in December 2023, there has not been any indication of government plans regarding Data and AI regulation or discussion. Recognizing this critical need, and at the request of the resident representative closely monitoring these issues, we are taking proactive steps to fill this gap.
In today's digital age, we generate an unprecedented amount of data through everyday activities such as using public transportation, making purchases, and interacting on social media. This surge in data production has vast potential to enhance various aspects of life, from improving web navigation and city management to monitoring pandemics and informing public policy. However, the mere availability of data is insufficient. To truly benefit from this wealth of information, it is crucial to know how to process, understand, and communicate it effectively, making ethical use of data and AI.
Our Lab actively began addressing this challenge last year. Since then, we have made progress on several actions. We have conducted interviews with a diverse group of stakeholders and experts across the country; from universities and foundations to laboratories and companies. These interviews provide valuable insights into the local AI landscape and our country's current position in AI advancement. This data forms the primary foundation for a report on the challenges and opportunities of AI for development in Argentina, scheduled for publication in the Q2 of 2024.
Our work strives to mainstream AI within the Country Office. This involves fostering conversations, engaging programme areas, and building staff capacity. To achieve this, we have shared key actions and invited senior management to participate in various activities. Additionally, we organized a training session utilizing AI tools for Country Office staff and consultants.
In addition to the aforementioned actions, we have also published three blogs discussing fundamental AI concepts, Argentinian public opinion on the technology and the debate on regulation. Alongside these blog entries, we published a report and two infographics showcasing the results of our 2023 pilot project integrating Natural Language Processing (NLP) into political speeches analysis.
Both the infographics and the report feature the findings of using a ChatGPT-based NLP model to analyze political speeches. This project aimed to identify shared issues, ideas and concerns within these speeches. As part of the anniversary of the 40 years of democracy in Argentina, leveraging AI tools, the Lab investigated presidential speeches delivered in Argentina since the return of democracy in 1983. The model successfully identified common topics and values among diverse speeches and established correlations between them.
This project served a dual purpose: as a proof of concept for AI’s potential in supporting democratic governance, and as a prototype to explore how language models can foster depolarization. For instance, the learnings could inform the application of these models in future research. Imagine using them to generate fictional text (vignettes) for survey experiments, evaluating voter's ability to distinguish between human-written and computer-generated text. This has significant relevance to studies on the proliferation of fake news during political campaigns.
The Lab’s focus on this topic is particularly timely. Deep political divisions and polarization, evident at both federal and state levels, significantly hinder progress on critical issues. This gridlock fuels public discontent erodes trust in governing institutions and fosters anti-establishment expressions. While overall democratic support remains high compared to other countries, these factors create a breeding ground for the erosion of civic values and governmental legitimacy. Dialogue and consensus building are urgently needed to bridge these divides and address pressing challenges.
To advance our goal of raising awareness and stimulating discussion about the potential and challenges of AI in development, we hosted a free, open seminar on the topic. The event convened experts, practitioners, and policymakers to explore AI’s potential in sectors like health, justice, government and business. Discussions addressed potential challenges and risks.
This well-attended event, drawing over 100 participants, marked a significant achievement in positioning UNDP as a key player in Argentina’s AI conversation. Further extending the reach of our work, Lorena Moscovich, our Head of Experimentation, delivered a TED Talk focused on AI-powered solutions for policy and development, to a capacity crowd of 11000 spectators. This presentation provided a valuable platform to share the insights and learnings garnered throughout 2023.
Summary of Results of the AI Agenda 2023:
TED Talk
Moscovich, L. (2024, February). Solutions for policy and development with AI [Video]. TED conferences. https://www.youtube.com/watch?v=-_0C5puviuc
Infographics
United Nations Development Programme [UNDP] (2024). The speech of democracy in its 40 years. https://inteligenciaargentina.org/discurso/en/
United Nations Development Programme [UNDP] (2023). Much in common. The topics and speech of democracy in its 40 years. An analysis of presidential speeches through artificial intelligence. https://inteligenciaargentina.org/en/
Report
Moscovich, L.; Vallejo Vera, S.; Alzú, M. S.; Gómez, F.; Zapata, M., and Moreno, M. V. (2023). Much in common. The topics and speech of democracy in its 40 years. An analysis of presidential speeches through artificial intelligence. Buenos Aires: UNDP. Available at: https://www.undp.org/es/argentina/publicaciones/report-ai-argentine-intelligence
Blogs
Zapata, M., Moscovich, L., Moreno, M.V. & López, M. E. (2024, June 15). Artificial Intelligence, Beyond Algorithms: An Evolving Regulatory Dilemma. Available at https://www.undp.org/es/argentina/blog/artificial-intelligence-beyond-algorithms-evolving-regulatory-dilemma
Moscovich, L. and De Rosa, A. (2023, October 9). Episode II. What do Argentines talk about when we talk about AI? (an example). UNDP Argentina. Available at https://www.undp.org/es/argentina/blog/episode-ii-what-do-argentines-talk-about-when-we-talk-about-ai-example
Moscovich, L. (2023, September 1). AI: Argentine Intelligence. A Conversation about Artificial Intelligence: Episode I: Anyone Can Cook, and Use AI! UNDP Argentina. Available at https://www.undp.org/es/argentina/blog/ai-argentine-intelligence-conversation-about-artificial-intelligence
Seminar
Seminar hosted by UNDP Argentina together with the MAPP of UdeSA in November 2023, titled: "Artificial Intelligence for Development" https://www.youtube.com/watch?v=iY2RqcFMqG8&t=15s
Poster at the Scientific Symposium
Moscovich, L.; Vallejo, S.; Zapata, M.; Alzú, M. S.; Gomez, F., and Moreno, M. V. (2023, October 4). Much in Common: NLP Applications for Depolarization Strategies. The Case of "The Discourse of Argentine Democracy in its 40 Years." Poster presented at the Scientific Symposium on AI and Applications, Buenos Aires, Argentina.
AI Training for the Country Office
In November 2023, a training session was facilitated on different Artificial Intelligence tools for all UNDP staff. The training was led by Dr. Tomás Balmaceda, a technology specialist.
Quantitative evidence: What (official) data sources do you have on this challenge that better exemplifies the importance and urgency of this frontier challenge? You can add text, a link, or a picture.
To address the rising global interest in how Data and AI impacts various aspects of development, we have initiated a conversation about this transformative technology. This conversation is crucial as the impact of big data and the influence of AI on people and states is multifaceted.
There is concern and uncertainty surrounding AI’s impact on the economy, particularly the labor market. While AI has the potential to significantly benefit the Latin America and Caribbean region through productivity gains (Neil Baily et al., 2023), it could also negatively impact employment.
The International Monetary Fund (IMF) predicted that AI will affect nearly 40% of jobs globally, replacing some roles and complementing others (Georgieva, 2024). The uncertainties surrounding AI's impact on developing countries add to the urgency of the challenge. At the beginning of the AI boom, the Inter-American Development Bank (IDB) estimated that between 36% and 43% of jobs in the region were at risk due to automation. Additionally, this outcome was not uniform across all countries; those with lower GDP per capita and higher inequality faced a greater risk of job loss (IDB, 2018). However, in addition to the loss of specific jobs, AI creates new precarious jobs: AI taggers. These workers are subcontracted by large technology companies to "train" AI systems, a tedious job, in exchange for low pay (BBC, 2023). Regardless of the exact figures, AI’s impact on employment and other related aspects, such as poverty and inequality, is undeniable.
AI also presents exciting possibilities for improving healthcare. It is already being used to accelerate drug development, analyze medical images with greater accuracy, and strengthen preventive medicine efforts (Yoon & Amadiegwu, 2023). In terms of efficiency, AI has the potential to generate savings of 5-10% in healthcare spending (Sahni, N. et al., 2023). However, ethical considerations remain a concern. Questions surround data privacy, bioethics, and the need to build patient trust. For instance, a 2023 Pew Research Center study found that 60% of adults in United States are uncomfortable with the idea of AI being used for medical treatments and diagnoses (Taylor, A. et al., 2023).
Similar to the ethical considerations, the regulation and governance of AI have become a central focus of discussion. A 2023 study by Stanford University analyzed legislative records from 127 countries and found a significant increase in AI-related legislation. The number of bills containing “artificial intelligence” and passed into law grew from just 1 in 2016 to 37 in 2022. This trend is further supported by an analysis of parliamentary records on AI in 81 countries, revealing a nearly 6.5-fold increase in mentions of AI in global legislative proceedings since 2016 (Stanford University, 2023).
The relationship between AI and political consensus is particularly crucial in the Latin American context. Data from Latinobarómetro, cited by the Economic Commission for Latin America and the Caribbean (CEPAL, by its acronym in Spanish), reveals a concerning trend in Argentina: declining public support for democracy and trust in governing institutions and political parties (CEPAL, n.d.).
Argentina’s political polarization is less focused on specific policy issues and more centered on political parties and leaders themselves. The data indicates that citizen’s support for policies can be conditional on the political party or leader endorsing them. This suggests a potential for consensus building around shared values and concerns.
However, a significant challenge remains. If people reject proposals based on their source (a specific politician or party), it will be difficult to overcome polarization. This highlights the need for a neutral mediator and facilitator of dialogue. This is where the UNDP sees an opportunity: to initiate these conversations using a tool not linked with partisan affiliations, such as AI.
Sources of data:
Comisión Económica para América Latina [CEPAL] (n.d.). Ficha países. https://www.cepal.org/es/analisis-la-inclusion-cohesion-social-america-latina-caribe-la-luz-pilar-social-la-agenda-2030-0
Georgieva, K. (2024, January 14). AI Will Transform the Global Economy. Let’s Make Sure It Benefits Humanity. IMF Blog. https://www.imf.org/en/Blogs/Articles/2024/01/14/ai-will-transform-the-global-economy-lets-make-sure-it-benefits-humanity
Inter-American Development Bank [IDB] (2018). Planet Algorithm: Artificial Intelligence for a Predictive and Inclusive form of Integration in Latin America. Integration and Trade Journal, 22(44). https://publications.iadb.org/en/integration-and-trade-journal-volume-22-no-44-july-2018-planet-algorithm-artificial-intelligence
Levy Yeyati, Moscovich, and Abuin (2020). Leader over policy? The scope of elite influence on policy preferences. Political Communication, 37(3), 398-422.
Neil Baily, M., Brynjolfsson, E. and Korinek, A. (2023). Machines of mind: The case for an AI-powered productivity boom. Brookings. https://www.brookings.edu/articles/machines-of-mind-the-case-for-an-ai-powered-productivity-boom/
Sahni, N; Stein, G.; Zemmel, R., and Cutler, D. (2023). What happens when AI comes to healthcare. Centre for Economic Policy Research [CEPR]. https://cepr.org/voxeu/columns/what-happens-when-ai-comes-healthcare
Smink, V. (2023, March 6). The hundreds of thousands of workers in poor countries that make artificial intelligence like ChatGPT possible (and why they generate controversy). BBC News Mundo. https://www.bbc.com/mundo/noticias-64827257
Tyson, A.; Pasquini, G.; Spencer, A. and Funk, C. (2023). 60% of Americans Would Be Uncomfortable with Provider Relying on AI in Their Own Health Care. Pew Research Center. https://www.pewresearch.org/science/2023/02/22/60-of-americans-would-be-uncomfortable-with-provider-relying-on-ai-in-their-own-health-care/
Yoon, S. and Amadiegwu, A. (2023). Emerging tech, like AI, is poised to make healthcare more accurate, accessible and sustainable. World Economic Forum [WEF]. https://www.weforum.org/agenda/2023/06/emerging-tech-like-ai-are-poised-to-make-healthcare-more-accurate-accessible-and-sustainable/
J. Boy, R. A. Rensink, E. Bertini and J. -D. Fekete, "A Principled Way of Assessing Visualization Literacy," in IEEE Transactions on Visualization and Computer Graphics, vol. 20, no. 12, pp. 1963-1972, 31 Dec. 2014, doi: 10.1109/TVCG.2014.2346984.
Borner, Katy & Bueckle, Andreas & Ginda, Michael. (2019). Data visualization literacy: Definitions, conceptual frameworks, exercises, and assessments. Proceedings of the National Academy of Sciences. 116. 201807180. 10.1073/pnas.1807180116.
Luvini, P.; Dias, J. M.; Kunst, M.; Ruiz Nicolini, J. P. y Yankelevich, D. (2023). Hacia un Estado Inteligente: una estrategia de datos para la Administración Pública Nacional. Fundar. Disponible en https://www.fund.ar
Qualitative evidence: What weak signals have you recently spotted that characterizes its urgency? Please provide qualitative information that better exemplifies the importance and urgency of this frontier challenge. You can add text, a link, or a picture.
Using data and AI responsibly and ethically offers significant potential to contribute to sustainable development and strengthen democratic societies. Insights gleaned from interviews with experts, alongside various sources like podcast, social media, blogs, and academic literature, reveal promising applications of data and AI across government, education, health, and justice.
For instance, the Fundar foundation is developing a data strategy for the national public administration. As a major producer of data, the State plays a crucial role in designing more effective public policies. However, the current fragmentation of data across various agencies limits its full potential. Addressing these gaps is essential for creating a smart State that can manage resources efficiently and support evidence-based policies.
Regarding AI, interviewees shared insights into its role in healthcare, including enhancing early disease detection, analyzing language cues for psychiatric conditions, and utilizing clinical records to identify public health disruptions. Additionally, Argentina's judiciary is exploring AI to improve the identification and characterization of rulings on gender-based violence. Optimizing public services and government procedures also emerged as a potential benefit of AI, as highlighted by interviewees.
However, to fully harness the potential benefits of Data and AI for democracy and sustainable development, we must acknowledge and mitigate its associated risks. Only by doing so can we maximize Data and AI’s benefits while minimizing its potential harms. Here are some key risks to consider:
- Concentration of Power: AI’s potential for data and power accumulation by certain actors poses a significant threat. This can undermine democratic principles and the rule of law by granting a select few undue influences over political and regulatory processes. Experts further warn that such concentration can stifle innovation. When research and development efforts align solely with the priorities of powerful AI actors, the field rises stagnation due to lack of diversity and fresh perspectives.
- Information Manipulation: AI can be used to manipulate information presented to voters, including the spread of fake news and misinformation.
- Algorithmic Bias: Implicit biases embedded in algorithms or in the data used to train them can lead to unfair and discriminatory outcomes, such as the exclusion of certain citizens groups. Experts have shared examples of AI used in parole decisions or government assistance programs exhibiting bias towards minorities or specific social groups.
- Difficulties in Oversight: The lack of transparency in the use of data and AI can limit citizens' ability to monitor and understand how these tools are being used and how policy decisions are being made. This opacity makes it challenging to contest decisions, as interviewees highlighted. Citizens cannot challenge an AI-made decision when the underlying mechanisms behind it are not clear.
- Privacy Threat: The vast amount of data collected by AI, particularly personal or sensitive data, raises concerns about potential violations of privacy, confidentiality, and image rights. Interviews revealed this as a significant challenge regarding AI use in healthcare and legal matters within Argentina.
- Geopolitics and AI: Experts worry that rising tensions between countries might slow down progress in AI research and development.
- Fallibility of AI: The implementation of AI systems, regardless of cause, is susceptible to errors. These errors can stem from technical malfunctions or human intervention. Interviewees emphasized the importance of adopting best practices from highly secure industries like aeronautical and nuclear power, where robust safety protocols are paramount.
Value proposition: What added value or unique value proposition is your Accelerator Lab bringing to solving this challenge? Why is it your Lab that needs to work on this challenge and not other actors within UNDP, other stakeholders in the country respectively? Why is it worth investing resources to this challenge?
The value proposition that our Accelerator Lab offers aligns with the values of the Digital 4 Development Hub in Latin America and the Caribbean. Specifically, we prioritize enhancing quality of life through technology and fostering citizen participation. By cultivating a people-centered digital culture, we aim to make public data more accessible and provide digital platforms for citizens to engage, ensuring that technology serves as a powerful tool for economic growth, creativity, and community building.
Also, the Lab, together with UNDP, are uniquely positioned to lead the public conversation on Data and AI use and regulation in the country, while promoting awareness of its benefits and risks for development and democracy. The UNDP’s strong reputation and history of convening stakeholders during critical moments make it a natural facilitator. For instance, during the 2001 economic and social crisis, UNDP played a central role in fostering dialogue and consensus-building around shared priorities. Today, while the context is different, UNDP can leverage its experience to identify common concerns and values that can serve as a foundation for a new national consensus on Data and AI, one that reflects the core priorities of the Argentine citizens. Thus, UNDP is a neutral mediator beyond the political divide.
The change in national administration presents an opportunity for the Lab to serve as a valuable partner. Given the limited experience with AI initiatives within the new government, the Lab’s knowledge and capabilities can help bridge potential gaps.
As a leader in innovation, we have been exploring various AI applications in areas like elections, democracy, and development. We actively share our findings with the Country Office and aid in enhancing its capacity to leverage the advantages of AI. Furthermore, the Lab has been successful in executing impactful actions, including the pilot project on NLP, the seminar hosted at the end of last year and the TED Talk facilitated by our Head of Experimentation, Lorena Moscovich.
This year, in the field of AI, there is the potential to hold a seminar in collaboration with the Ministry of Foreign Affairs and to present a summary of our AI research findings at Universidad Siglo XXI in Córdoba. In the area of data and digitalization, we have been selected to present at the UN World Data Forum on 'PreserVamos' and our experience in generating environmental data collaboratively. Additionally, we will organize the 'Contar con Datos' contest in partnership with the Data Science Lab of the University of San Andrés, in collaboration with the Undersecretariat of Science of the Nation. This data visualization contest aims to expand the data community and promote collaborative data exploration.
Short “tweet” summary: We would like to tweet what you are working on, can you summarize your challenge in a maximum of 280 characters?
Join us to discuss #Data and #AI use and regulation in Argentina. Can it boost sustainable development? Are there risks? We use #CollectiveIntelligence and #DataVisualization to turn daily data into insights, driving evidence-based decisions and fostering public conversation.
Learning questions
Learning question: What is your learning question for this challenge? What do you need to know or understand to work on your challenge statement?
Learning questions that arise from our challenge:
- What are the elements, process, institutions and models involved in the regulation of AI worldwide? Is it feasible to replicate these in Argentina? Has the country its own laws and experiences on AI?
- What are the characteristics of AI usage and access in Argentina? Are these compatible with ethical, sustainable, and responsible standards?
- Who are the main players within the national AI ecosystem?
- Can a data visualization contest serve as a tool for collective intelligence to explore publicly available data in Argentina and identify positive deviants?
- What are the community’s areas of interest regarding data?
- Can such a contest help foster data appropriation and enhance data visualization literacy?
To meet this challenge and its related learning questions, the Lab has drafted a portfolio of actions. Within the second quarter of the year, we plan to publish a report on the challenges and opportunities of AI for development in Argentina. This report will synthesize what the Lab has learnt from around 25 interviews conducted with key AI experts.
In addition to this report, this year we published a blog post that analyzes the current state of AI-related laws and regulations in Argentina. We are also organizing a new seminar to be delivered this year in collaboration with the Ministry of Foreign Affairs, focusing on regulatory topics. The report, the blog entry, and the seminar will join the lineup of products and events we rolled out in 2023, which can be checked out in the sections above.
Regarding data and digitalization, we will present a summary of our AI research findings at Universidad Siglo XXI in Córdoba. Also, we have been selected to present at the UN World Data Forum on 'PreserVamos' and our experience in generating environmental data collaboratively. Additionally, we will organize the 'Contar con Datos' contest in partnership with the Data Science Lab of the University of San Andrés, in collaboration with the Undersecretariat of Science of the Nation.
The Lab will further undertake a prospective analysis to identify signals of change that allow us to explore emerging trends and development regarding the AI for Development Agenda. This analysis will encompass the exploration systematic examination of relevant signals.
To what stage(s) in the learning cycle does your learning question relate?
Sense, Explore, Test
Usage of methods: Relating to your choice above, how will you use your methods & tools for this learning question? What value do these add in answering your learning question?
The action plan on AI, data and digitalization employs a comprehensive methodology, including stakeholders mapping, interviews with key informants, blogs publishing, application of AI models, collective data generation, data visualization, data contest and the convening of seminars. The long-term ambition of this multi-pronged approach is to leverage collective intelligence to foster a deeper understanding of how to use and regulate data and AI, ensuring a balanced approach that maximizes its potential benefits while mitigating associated risks.
Furthermore, the Lab’s exploration of AI applications within the context of political discourse extends beyond a mere proof-of-concept. It serves as a prototype for testing the potential of AI to contribute to strengthen democratic governance. As previously mentioned, a Natural Language Processing (NLP) model using Chat GPT was developed to analyze political speeches. This model successfully identified recurring themes and shared values among diverse speeches, establishing correlative relationships. Additionally, the model processed real political speeches to generate artificial ones. The pilot project serves as a compelling demonstration of AI’s potential to identify unifying elements within a nation’s political history, showcasing a responsible application of this technology that aligns with democratic principles.
The pilot project holds significant value for further research on the dissemination of misinformation and “fake news” within democratic environments. The extracted insights could facilitate the application of these models in generating fictitious text for survey-based experiments (vignettes) and assessing a voter's capacity to discern between human-generated and computer-generated text.
Furthermore, the Lab leverages AI-powered analytics to examine social media data, aiming to understand public sentiment and opinions on AI within Argentina. A collaborative effort with Citibeats, a social listening company, resulted in the analysis of nearly 22 million tweets from public accounts georeferenced in Argentina during April,2023. This data provides valuable insights into public perception.
The Lab’s approach to fostering collective intelligence involves in-depth interviews with a diverse pool of experts on the cutting-edge applications and regulations of AI for both democratic governance and development. These interviews facilitate the gathering of insights and case studies regarding the potential benefits and risks associated with AI. The knowledge acquired serves as a foundation for establishing categories and variables for analyzing AI aspects relevant to the Lab’s core challenge.
Finally, the Lab hosted a seminar on AI that fostered open exchange with a broad audience. Experts and over 100 attendees engaged in discussions on approaches and models for regulating AI and its impact on development.
Existing data gaps: Relating to your choice above, what existing gaps in data or information do these new sources of data addressing? What value do these add in answering your learning question?
Discussions regarding the ethical consideration surrounding the use of data, as well as the development and deployment of AI are ongoing. However, these discussions often lack focus on critical aspects, such as:
Promoting digital literacy and the ability to interpret data.
Identifying key players and stakeholders within the national AI ecosystem, including their respective roles and agendas.
Examining potential regulatory models that can support government authorities in fostering ethical, responsible, and secure data use and AI development
Analyzing the pressing challenges that AI presents for Argentina, particularly in the context of its development efforts
Learning questions
Learning question: What is your learning question for this challenge? What do you need to know or understand to work on your challenge statement?
Learning questions that arise from our challenge:
- What are the elements, process, institutions and models involved in the regulation of AI worldwide? Is it feasible to replicate these in Argentina? Has the country its own laws and experiences on AI?
- What are the characteristics of AI usage and access in Argentina? Are these compatible with ethical, sustainable, and responsible standards?
- Who are the main players within the national AI ecosystem?
- Can a data visualization contest serve as a tool for collective intelligence to explore publicly available data in Argentina and identify positive deviants?
- What are the community’s areas of interest regarding data?
- Can such a contest help foster data appropriation and enhance data visualization literacy?
To meet this challenge and its related learning questions, the Lab has drafted a portfolio of actions. Within the second quarter of the year, we plan to publish a report on the challenges and opportunities of AI for development in Argentina. This report will synthesize what the Lab has learnt from around 25 interviews conducted with key AI experts.
In addition to this report, this year we published a blog post that analyzes the current state of AI-related laws and regulations in Argentina. We are also organizing a new seminar to be delivered this year in collaboration with the Ministry of Foreign Affairs, focusing on regulatory topics. The report, the blog entry, and the seminar will join the lineup of products and events we rolled out in 2023, which can be checked out in the sections above.
Regarding data and digitalization, we will present a summary of our AI research findings at Universidad Siglo XXI in Córdoba. Also, we have been selected to present at the UN World Data Forum on 'PreserVamos' and our experience in generating environmental data collaboratively. Additionally, we will organize the 'Contar con Datos' contest in partnership with the Data Science Lab of the University of San Andrés, in collaboration with the Undersecretariat of Science of the Nation.
The Lab will further undertake a prospective analysis to identify signals of change that allow us to explore emerging trends and development regarding the AI for Development Agenda. This analysis will encompass the exploration systematic examination of relevant signals.
To what stage(s) in the learning cycle does your learning question relate?
Sense, Explore, Test
Usage of methods: Relating to your choice above, how will you use your methods & tools for this learning question? What value do these add in answering your learning question?
The action plan on AI, data and digitalization employs a comprehensive methodology, including stakeholders mapping, interviews with key informants, blogs publishing, application of AI models, collective data generation, data visualization, data contest and the convening of seminars. The long-term ambition of this multi-pronged approach is to leverage collective intelligence to foster a deeper understanding of how to use and regulate data and AI, ensuring a balanced approach that maximizes its potential benefits while mitigating associated risks.
Furthermore, the Lab’s exploration of AI applications within the context of political discourse extends beyond a mere proof-of-concept. It serves as a prototype for testing the potential of AI to contribute to strengthen democratic governance. As previously mentioned, a Natural Language Processing (NLP) model using Chat GPT was developed to analyze political speeches. This model successfully identified recurring themes and shared values among diverse speeches, establishing correlative relationships. Additionally, the model processed real political speeches to generate artificial ones. The pilot project serves as a compelling demonstration of AI’s potential to identify unifying elements within a nation’s political history, showcasing a responsible application of this technology that aligns with democratic principles.
The pilot project holds significant value for further research on the dissemination of misinformation and “fake news” within democratic environments. The extracted insights could facilitate the application of these models in generating fictitious text for survey-based experiments (vignettes) and assessing a voter's capacity to discern between human-generated and computer-generated text.
Furthermore, the Lab leverages AI-powered analytics to examine social media data, aiming to understand public sentiment and opinions on AI within Argentina. A collaborative effort with Citibeats, a social listening company, resulted in the analysis of nearly 22 million tweets from public accounts georeferenced in Argentina during April,2023. This data provides valuable insights into public perception.
The Lab’s approach to fostering collective intelligence involves in-depth interviews with a diverse pool of experts on the cutting-edge applications and regulations of AI for both democratic governance and development. These interviews facilitate the gathering of insights and case studies regarding the potential benefits and risks associated with AI. The knowledge acquired serves as a foundation for establishing categories and variables for analyzing AI aspects relevant to the Lab’s core challenge.
Finally, the Lab hosted a seminar on AI that fostered open exchange with a broad audience. Experts and over 100 attendees engaged in discussions on approaches and models for regulating AI and its impact on development.
Existing data gaps: Relating to your choice above, what existing gaps in data or information do these new sources of data addressing? What value do these add in answering your learning question?
Discussions regarding the ethical consideration surrounding the use of data, as well as the development and deployment of AI are ongoing. However, these discussions often lack focus on critical aspects, such as:
Promoting digital literacy and the ability to interpret data.
Identifying key players and stakeholders within the national AI ecosystem, including their respective roles and agendas.
Examining potential regulatory models that can support government authorities in fostering ethical, responsible, and secure data use and AI development
Analyzing the pressing challenges that AI presents for Argentina, particularly in the context of its development efforts
11Sustainable cities and communities
16Peace, justice and strong institutions
Comments
Log in to add a comment or reply.