he rapid expansion of artificial intelligence (AI) has structurally transformed decision-making processes in areas such as healthcare, education, the public sector, and environmental management. Far from being a neutral set of technical tools, AI today constitutes a sociotechnical infrastructure that embeds values, normative assumptions, and power relations. In this context, the Artificial Intelligence for Social Good (AI4SG) approach emerges as a theoretical–practical framework aimed at aligning the design, implementation, and evaluation of AI systems with explicit goals of social well-being, equity, and sustainability.
AI4SG (Artificial Intelligence for Social Good) can be defined as a field of research and practice that seeks to apply advances in artificial intelligence to address social problems and improve the well-being of individuals, society, and the planet as a whole.
AI4SG is not limited to the application of advanced technologies to social challenges; rather, it proposes a normative reorientation of algorithmic innovation, integrating applied ethics, governance, and social impact assessment as structural components of technological development.
From Technological Optimism to Algorithmic Critique
Contemporary critical literature has shown that the indiscriminate adoption of algorithmic systems can generate significant adverse effects. O’Neil (2016) documents how opaque predictive models, even when statistically robust, can amplify inequalities and consolidate forms of structural exclusion. Complementarily, Benjamin (2019) demonstrates that AI systems tend to reproduce pre-existing racial and social hierarchies, configuring what she calls the New Jim Code.
From a broader perspective, AI is increasingly understood as a material and political phenomenon, whose operation depends on global chains of resource extraction, precarious human labor, and asymmetric concentrations of power (Crawford, 2021). These contributions converge in highlighting that technical efficiency is not a sufficient criterion for assessing the social legitimacy of AI, thereby opening conceptual space for approaches such as AI4SG.
Definition and Scope of AI4SG
Following Cowls (2022), AI4SG can be defined as the set of approaches, methodologies, and practices aimed at maximizing the positive social impact of AI while minimizing ethical, social, and environmental risks. This approach is characterized by three fundamental features:
Normative intentionality: social objectives are not collateral effects, but explicit goals of the system.
Human-centeredness: AI is conceived as support for human deliberation and decision-making, not as a substitute for moral responsibility.
Impact assessment: system performance is measured in both technical and social terms.
From this perspective, AI4SG lies at the intersection of data science, technology ethics, and public policy.
The specialized literature converges around a set of principles that structure AI4SG projects:
Algorithmic justice and equity, through the identification and mitigation of bias.
Transparency and explainability, as conditions for public trust.
Responsibility and accountability, clearly defining actors and roles.
Precaution and proportionality, especially in contexts of high vulnerability.
Verifiable social impact, beyond operational efficiency.
Christian (2020) conceptualizes this challenge as the alignment problem, emphasizing that aligning intelligent systems with human values is simultaneously a technical, institutional, and moral problem.
AI4SG in Critical Sectors: The Case of Healthcare and the Public Sector
I can be deployed across multiple domains to positively impact individuals, communities, or ecosystems:
Social inclusion: helping reduce gaps through applications that facilitate communication for people with disabilities or tools that detect gender bias in hiring and credit processes.
Health and well-being: used to diagnose diseases (such as sepsis or diabetic retinopathy) through the analysis of medical records and images. It also enables telemedicine, allowing healthcare services to reach remote areas via mobile devices.
Quality education: enabling personalized learning systems (intelligent tutors or avatars) that adapt to the pace and specific needs of each learner.
Agriculture and the environment: applied in precision agriculture through robots that optimize planting and irrigation, as well as in climate monitoring, marine life protection, and anti-poaching efforts using drones and computer vision algorithms.
Similarly, in the public sector, the application of AI to the targeting of social policies requires robust governance frameworks to prevent the uncritical automation of decisions with high social impact.
Decolonial Analysis of AI4SG: Epistemic Limits and Emancipatory Possibilities
From a decolonial perspective, the AI4SG approach requires additional problematization that goes beyond dominant normative frameworks in AI ethics. Following Aníbal Quijano, modern technology is inseparably linked to the coloniality of power, understood as a historical pattern that articulates knowledge, economy, and authority. In this sense, AI—including that oriented toward the “social good”—cannot be assumed to be neutral or universal.
Most AI systems are designed based on epistemologies and technical rationalities rooted in the Global North, implying that social problems and optimization criteria are often defined from frameworks external to the contexts in which these technologies are deployed. As Walter Mignolo warns, this process reproduces a form of coloniality of knowledge, whereby certain forms of knowledge are legitimized as universal while others are systematically marginalized.
Artificial Intelligence for Social Good currently constitutes an indispensable framework for guiding the development of artificial intelligence in contexts of high social complexity. By integrating ethics, governance, and epistemological critique, AI4SG makes it possible to move beyond reductionist views of technological innovation and advance toward a conception of AI as a tool in the service of collective well-being. The incorporation of a decolonial perspective further expands this approach, reminding us that there can be no true “social good” without epistemic justice, cultural contextualization, and the effective participation of affected communities. Ultimately, the future of AI4SG will depend on institutional capacity to translate these principles into concrete practices of design, regulation, and evaluation.
References
Benjamin, R. (2019). Race after technology: Abolitionist tools for the new Jim code. Polity Press. Christian, B. (2020). The alignment problem: Machine learning and human values. W. W. Norton & Company. Coeckelbergh, M. (2020). AI ethics. MIT Press. Cowls, J. (Ed.). (2022). Artificial intelligence for social good. Springer. Crawford, K. (2021). Atlas of AI: Power, politics, and the planetary costs of artificial intelligence. Yale University Press. O’Neil, C. (2016). Weapons of math destruction: How big data increases inequality and threatens democracy. Crown.
Higher Education at the Convergence of Artificial Intelligence, Demographic Transition, and Generational Change
Higher education is undergoing a structural reconfiguration driven by three converging dynamics: the expansion of artificial intelligence (AI) and digital hyperconnectivity, demographic decline across multiple regions, and the transformation of cultural and professional expectations among younger generations.
This text reflects on these forces, which compel us to rethink three fundamental strategic questions — Who should we educate? What should we educate about? Where should we educate? — not as isolated pedagogical concerns, but as axes for institutional redesign.
Based on evidence from international organizations and recent academic literature, it is argued that the sustainability and relevance of higher education will depend on its ability to adopt an intergenerational model, integrate digital literacy transversally, and evolve toward a hybrid educational architecture.
From Mass Expansion to Structural Reconfiguration Over the past decades, higher education experienced unprecedented expansion. According to UNESCO (2022), global enrollment increased from approximately 100 million students in 2000 to more than 235 million in 2020. However, this expansion developed under demographic and technological assumptions that are now being profoundly altered.
The advancement of artificial intelligence is reshaping work dynamics, learning processes, and knowledge production. UNESCO (2023) warns that AI impacts not only pedagogical tools but also the very architecture of educational systems, including assessment, curriculum design, and institutional governance.
Simultaneously, the United Nations Population Division reports that the global fertility rate declined to 2.3 children per woman in 2021, projected to fall below replacement level in multiple regions (United Nations, 2022). This phenomenon directly affects the size of university-age cohorts that have traditionally sustained higher education systems.
Moreover, the generations currently entering educational processes are themselves undergoing transformation. Research on Generation Z shows significant shifts in expectations regarding learning and work. Twenge (2017) documents changes in digital socialization patterns and career priorities, while Gallup (2024) highlights ambivalent perceptions regarding AI’s impact on education and employment.
Generational transformation extends beyond pedagogical preferences; it reshapes the relationship between education and life planning. Institutions that fail to integrate flexibility, modularity, applied learning, and professional transition support risk losing relevance among students who value both employability and ethical coherence in their educational experience.
In this context, higher education is not facing a temporary crisis, but a systemic redefinition.
Who Should We Educate? Demographic Transition and the Shift Toward Lifelong Learning
Demographic transition represents one of the most structural drivers of change. The World Population Prospects report (United Nations, 2022) confirms a sustained slowdown in youth population growth across Latin America, Europe, and East Asia.
Encoura (2023) projects a significant decline in high school graduates in the United States over the next decade — a phenomenon mirrored in other countries with reduced fertility rates.
This scenario transforms institutional logic:
Fewer traditional students
Increased interinstitutional competition
The need to diversify target populations
The OECD (2019) argues that lifelong learning will be critical for sustaining productivity in aging societies. Consequently, higher education must expand its focus toward:
Professionals undergoing reskilling due to automation
The traditional student is no longer the exclusive center of the model. Institutional sustainability will increasingly depend on the ability to operate as a lifelong learning platform.
What Should We Educate About? AI, Automation, and Expanded Human Competencies
Artificial intelligence is redefining labor market competencies. The Future of Jobs Report (World Economic Forum, 2023) identifies technological literacy, data analysis, and analytical thinking as rapidly growing skills.
Brynjolfsson and McAfee (2014) argue that automation transforms not only manual tasks but also cognitive ones, requiring the reconfiguration of professional profiles. UNESCO (2023) emphasizes that responsible AI integration in education must include ethical frameworks, algorithmic transparency, and critical thinking development.
In this context, transversal training in:
Data literacy
Applied AI
Automation
Digital ethics
becomes a foundational requirement rather than an optional specialization.
At the same time, literature consistently highlights that automation does not eliminate human value; rather, it shifts it toward higher-order competencies. Technology complements tasks requiring judgment, creativity, and social intelligence.
Higher education must therefore cultivate hybrid professionals: technologically competent and strong in advanced cognitive capacities.
Where Should We Educate? Hyperconnectivity and the Expansion of the Learning Ecosystem
Hyperconnectivity structurally redefines educational space. Castells (2010) describes the “network society” as a system in which knowledge production, circulation, and validation occur through global digital infrastructures. Information is no longer confined to closed institutions but flows continuously within interconnected networks.
Universities, historically organized as centralized physical spaces, now operate within a distributed architecture of knowledge.
The COVID-19 pandemic accelerated virtualization processes but did not initiate them. Educational digitalization was already advancing through open learning platforms, digital educational resources, online communities of practice, and global professional networks. The pandemic revealed the fragility of exclusively face-to-face models and the necessity of hybrid institutional capabilities.
Yet the transformation goes beyond migration to virtual environments. Contemporary learning occurs simultaneously across multiple spaces:
Open learning platforms
Corporate training environments
Specialized digital communities
Global professional networks
Innovation and entrepreneurship ecosystems
Self-directed learning
Within this context, the emerging higher education model displays distinctive characteristics:
Hybrid education. Not merely a technical blend of in-person and online formats, but an integrated pedagogical design combining physical, digital, and experiential learning.
Microcredentials. The OECD (2021) notes that microcredentials certify specific, updatable competencies, enabling flexible and stackable learning pathways that respond to both lifelong learning needs and employer demand for verifiable skills.
Competency-based assessment. Emphasis shifts from credit accumulation to demonstrable learning outcomes. Certification validates mastery rather than time spent.
Integration with real productive environments. Learning connects with business projects, living labs, simulations, extended internships, and sectoral challenges. The boundary between classroom and market becomes increasingly porous.
This redesign also responds to cultural transformation. Generation Z values flexibility, purpose, and immediate applicability of learning (Gallup, 2024). It seeks personalized pathways, relevant experiences, and direct connections between education and employability. Rigid curricular structures lose appeal compared to adaptive, modular models.
The physical campus does not disappear. However, its function is redefined. It ceases to be the sole node of the educational system and becomes one component within an expanded learning network. It evolves from exclusive container of knowledge to space for encounter, experimentation, social interaction, and academic community building.
The university of the 21st century is defined not solely by its physical infrastructure, but by its capacity to articulate hybrid ecosystems, digital networks, and productive environments into a coherent and strategic learning experience.
Final Reflection
The convergence of demographic transition, cognitive automation, and generational cultural transformation is shaping an environment that is structurally different from the one that gave rise to the traditional university model.
We are not facing marginal adjustments, but rather a fundamental alteration of the system’s foundational assumptions: the abundance of young populations, the stability of professional profiles, and the centrality of the physical campus as the sole legitimate space for education.
The emerging scenario is clear: Fewer young students. More transversal technology. Growing expectations for flexibility, purpose, and immediate applicability.
This context redefines competition in higher education. It is no longer sufficient to expand coverage, diversify programs, or strengthen infrastructure alone. The challenge is to rethink the institutional value proposition in terms of relevance, adaptability, and the capacity to articulate effectively with a dynamic environment.
Institutions that understand the systemic nature of this convergence will be positioned to redesign their academic architecture, diversify their target populations, and consolidate themselves as lifelong learning platforms. Those that maintain exclusively expansion-driven logics — centered on volume, rigid presenciality, or closed disciplinary segmentation — will face increasing pressures on their long-term sustainability.
Referencias
Autor, D. H. (2015). Why are there still so many jobs? Journal of Economic Perspectives, 29(3), 3–30. Brynjolfsson, E., & McAfee, A. (2014). The second machine age. W. W. Norton. Castells, M. (2010). The rise of the network society. Wiley-Blackwell. Encoura. (2023). Regional impacts of the demographic decline on higher education. Gallup. (2024). Gen Z and AI in education. OECD. (2019). Getting skills right: Future-ready adult learning systems. OECD. (2021). Micro-credentials for lifelong learning and employability. UNESCO. (2022). Global education monitoring report. UNESCO. (2023). AI and the future of education: Disruptions, dilemmas and directions. United Nations. (2022). World population prospects 2022. World Economic Forum. (2023). Future of jobs report 2023.
Latin America’s rapid urbanization has triggered profound challenges: escalating pollution, overpopulation, inefficient transportation systems, and a scarcity of green spaces. These problems, intensified by social, economic, and infrastructural limitations, necessitate a thorough overhaul of urban design to promote sustainability and resilience. In this critical context, innovative solutions are vital to reshape the region’s future metropolises. Belgian architect Vincent Callebaut provides a visionary response, merging technology, ecology, and urban planning to develop smart, sustainable cities. His proposals—self-sufficient buildings, vertical farms, and renewable energy systems—offer a foundation for a potential urban revolution in Latin America. This prompts the question: How can these ideas be tailored to the region’s distinct realities? This article explores Callebaut’s concepts and assesses their capacity to create more livable, sustainable urban environments in Latin America.
Vincent Callebaut, born in 1977, is a distinguished architect celebrated for his dedication to ecological architecture and sustainable urbanism. A graduate of the Victor Horta Institute of Architecture in Brussels, he has pioneered projects that seamlessly blend technology, nature, and futuristic design. His philosophy combines ecological architecture, sustainable urban planning, and biomimicry, focusing on resilient, self-sufficient urban centers. Utilizing state-of-the-art technologies, renewable energy, and sustainable materials, his designs aim to transform cities into vibrant, eco-friendly spaces. His numerous awards—Green Practitioner of the Year 2021, Best Execution Architecture 2023, and the Global Quality Silver Pyramid 2024—affirm his standing as a global leader in sustainable architecture.
For Callebaut, design and aesthetics play a central role in his work—not just as visual elements, but as expressions of a conceptual vision that integrates functionality, innovation, and beauty. His approach is distinguished by meticulous attention to detail and a seamless harmony between form and purpose, allowing his projects to transcend mere utility and become manifestations of a broader vision for the future. In this sense, his futuristic perspective is evident in his use of cutting-edge materials, emerging technologies, and concepts that anticipate evolving social, cultural, and environmental dynamics. Through his work, he not only envisions possible scenarios but brings them to life through designs that challenge conventional boundaries and propose new ways of interacting with the human environment.
Among his most representative works is the Taijitu project, designed in 2024—a sustainable sports center dedicated to the practice of Tai Chi Chuan. Located in Shenyang, China, on the banks of the Hunhe River, this 4,750 m² complex blends harmoniously with its natural surroundings. Inspired by the Yin-Yang symbol, its biomimetic architecture adopts a double-spiral form that reinterprets traditional curved wooden roof structures, adhering to the principles of balance and symmetry inherent in Chinese culture. The design reflects Callebaut’s philosophy, which seamlessly integrates sustainability, biomimicry, and advanced technology to create highly innovative projects.
The Dune project (2025), in turn, is an innovative biomimetic architecture proposal that merges urbanism, ecology, and technology to create a self-sufficient and climate-resilient environment. Inspired by the organic forms of dunes and coastal ecosystems, this design incorporates sustainable materials, renewable energy, and natural ventilation systems to optimize resource consumption. With a structure designed to support environmental regeneration, Dune seeks to redefine the relationship between cities and nature, promoting an urban habitat model in harmony with the planet.
These projects open a pathway for reflection on housing construction in Latin America. The use of recycled materials and innovative techniques, such as self-healing bioconcrete, could provide viable alternatives for developing more durable and sustainable social housing in the region. Likewise, the incorporation of renewable energy, green roofs, and urban farms could enhance the quality of life in densely populated areas. A biomimetic and energy self-sufficiency approach would allow Latin American cities not only to expand but to do so in harmony with their surroundings, simultaneously addressing environmental and social challenges.
Beyond individual architecture, Callebaut envisions large-scale projects that transform urban planning with a futuristic perspective and an ethic of harmony with nature. A notable example is the Nautilus Eco-Resort, located in Palawan, Philippines. Designed as a biomimetic learning center with zero emissions, zero waste, and zero poverty, this sustainable complex merges ecological architecture with responsible tourism. Featuring 12 spiral towers and modular structures, it integrates renewable technologies such as solar and wind energy, along with water and waste recycling systems. This design aims to foster biodiversity and environmental education, offering a model of self-sufficient and environmentally respectful development.
In the Latin American context, these types of designs could have a transformative impact on coastal regions vulnerable to climate change and environmental degradation. Biomimetic proposals that integrate renewable energy and recycling could help mitigate the effects of mass tourism and uncontrolled growth in the Mexican Caribbean, the Colombian coasts, and Central American islands. Moreover, by generating green jobs and fostering resilience to extreme climate events, these initiatives would provide solutions tailored to local needs.
Along the same lines of marine- and ocean-focused innovation, the Lilypad project represents a groundbreaking proposal for a sustainable floating city designed to address the challenges of climate change and rising sea levels. Inspired by the shape of a lotus flower, this city aims to be a self-sufficient refuge for climate refugees, providing housing, community spaces, and agricultural areas. Its design incorporates advanced ecological technologies, including energy generation from solar, wind, and geothermal sources, as well as seawater desalination for potable water supply. The structure is designed to adapt to aquatic environments, featuring floating platforms capable of adjusting to fluctuations in water levels.
This project embodies Callebaut’s vision of a future where cities not only adapt to environmental conditions but also offer innovative solutions to the global climate crisis. In the Latin American context, particularly in coastal regions vulnerable to rising sea levels and the impacts of climate change, Lilypad could present a viable alternative. Countries such as Colombia, Mexico, Peru, and the Caribbean island nations face severe risks due to sea level rise and the intensification of extreme weather events. The possibility of floating, self-sufficient living spaces could help alleviate pressure on densely populated urban areas, mitigating the adverse effects of uncontrolled urbanization in these regions.
The Dragonfly project presents a futuristic concept of a sustainable skyscraper, inspired by nature and designed to transform the urban landscape through innovative and eco-friendly architecture. Its structure, reminiscent of a dragonfly’s form, features wings engineered to maximize the capture of solar and wind energy, enabling the building to operate self-sufficiently. Additionally, it integrates green technologies such as solar panels, wind turbines, and water recycling systems, creating an autonomous urban ecosystem that accommodates residential, commercial, and vertical farming spaces.
With this proposal, Callebaut envisions a future in which buildings not only fulfill conventional functions but also actively contribute to the regeneration of the urban environment. The application of such projects in major Latin American cities could have a significant impact, particularly in São Paulo, Mexico City, Buenos Aires, and Bogotá, where rapid urban growth has led to critical issues such as pollution, traffic congestion, and resource scarcity. The incorporation of sustainable urban models could provide a viable solution to these challenges. Integrating renewable energy, recycling systems, and urban agriculture into contemporary architecture would help reduce cities’ ecological footprints and improve the quality of life for their inhabitants.
This vision of urban transformation is also reflected in other projects by Callebaut, such as Hyperion and Paris Smart City 2050. Hyperion is an ecological skyscraper concept inspired by the form of a tree, designed to integrate renewable energy and water recycling systems to generate a positive impact on both the environment and urban life.
Meanwhile, Paris Smart City 2050 is a visionary model of a smart and sustainable city that combines advanced architecture, intelligent mobility, efficient resource management, and the integration of green spaces to create resilient and self-sufficient urban environments in the face of climate change challenges. Both projects represent a synthesis of technology, nature, and urban design, shaping a paradigm of sustainable cities for the future.
In this context, the Paris Smart City 2050 project could serve as a reference for the modernization of cities in Latin America. In a region characterized by rapid urban growth, high levels of inequality, and environmental challenges such as pollution and deforestation, Callebaut’s vision offers adaptable solutions. His proposals include self-sufficient buildings powered by solar or wind energy, the integration of green spaces to mitigate urban heat, and local food production to reduce dependence on external supply chains.
Megacities such as Bogotá, Mexico City, and São Paulo could benefit from these ideas to ease pressure on resources, improve air quality, and promote a circular economy. The key lies in adapting these designs to local conditions, addressing specific challenges such as tropical rainfall management and the use of regional materials and labor, ensuring a balance between innovation and cultural context.
Another of his most recent works is Écume des Ondes (2024) in Aix-les-Bains, France, a transformation of the old thermal baths into a sustainable wellness center with undulating green terraces and an aquaponic farm. Flower Tower (2024) in Brussels is a hybrid wooden hospital that prioritizes biophilic design, while Harmocracy (2024) in Neom, Saudi Arabia, is a futuristic airport that optimizes solar energy and natural ventilation.
Innovative proposals also include Green Line (2023) in Geneva, a car-free eco-district with cascading wooden villas, and Green New Deal (2023) in New York, which reimagines the city with vertical villages designed to reduce emissions by 85% by 2050. In 2022, Manta Ray (Seoul) transformed a disused highway into a productive space with agricultural bridges, while Archibiotec (Paris) created an urban distillery that converts waste into biofuels.
Meanwhile, Pollinator Park (2020), commissioned by the European Commission, is a virtual park that educates visitors on the importance of pollinators through organic structures simulating natural ecosystems. Conceptual projects such as Hydrogenase (2008) envision zero-emission airships powered by biohydrogen from algae, while Physalia (2007) is a floating garden vessel designed to purify European rivers using solar energy and biofiltration. Perfumed Jungle (2006) in Hong Kong transforms the waterfront into a «green lung» with interwoven ecological towers. Anti-Smog (2005) in Paris, France, is an ecological center that purifies the air through green technologies and biomimetic design. Lastly, Elasticity (2001), an academic project, proposed an autonomous aquatic city for 50,000 people, marking the beginning of Callebaut’s futuristic vision.
However, implementing these designs in Latin America presents multiple challenges. The uncontrolled expansion of cities and the proliferation of informal settlements complicate sustainable infrastructure planning. Additionally, economic inequality limits equitable access to ecological technologies. Environmental conditions—such as air pollution, deforestation, and climate variability—require specific architectural adaptations to optimize resource management across diverse climates. The integration of a circular economy and the use of local materials can help reduce costs and promote regional employment, but their implementation necessitates structural changes in production and consumption models. Finally, technological infrastructure and community training are essential to ensuring the long-term sustainability of these projects.
Consequently, a comprehensive approach is required—one that combines innovation, inclusive public policies, and contextual strategies that address the specific needs of each region.
Callebaut’s urban vision offers an innovative framework for rethinking the design of future cities, integrating technology, ecology, and functionality into a model of sustainable urbanism. His proposals not only anticipate the challenges of climate change and urban expansion but also present viable solutions to reduce environmental impact and improve quality of life in densely populated environments. In the Latin American context, where cities face structural issues such as inequality, pollution, and infrastructure deficits, adapting these concepts is crucial. However, their implementation demands coordinated public policies, technological investment, and a shift in urban planning that prioritizes sustainability and resilience.
Beyond aesthetics and architectural innovation, the real challenge lies in transforming these ideas into tangible, accessible solutions tailored to local realities. The future of cities will depend on the ability to merge vision with action, adopting strategies that enable a balanced development between urban growth and environmental preservation.
Note: I am deeply grateful to Professor Carolina Espitia for highlighting the importance of climate change awareness through the Latin American Chair of Environmental Thought and Climate Crisis at Universidad Central. Her guidance has inspired me to reflect and take action toward a sustainable future.
