Education and public policy experts weigh in on the educational, cultural, and environmental implications of AI, putting emphasis on Indigenous knowledge, data sovereignty, and cultural integrity.
In photo (from left to right): Prof Michael Menchaca (The University of Hawai’i at Mānoa, United States), IAFOR President Prof Jun Arima (The University of Tokyo, Japan), Prof Peter J. Mataira (Hawaii Pacific University, United States), and Dr LJ Rayphand (Caroline College and Pastoral Institute in Chuuk State, Federated States of Micronesia) during their panel presentation at IICE/IICAH2026
While the arts and humanities are severely underfunded, some areas within education have experienced growth and have been permitted to operate unregulated. Artificial Intelligence (AI) in particular has witnessed exponential growth and interest at universities and beyond, both within research and application. However, this growth does not come without risks, endangering not only academic integrity and the future of employment, but also what it means to be human and what constitutes human intelligence. Indigenous people, knowledge, and the environment are further casualties of the exponential, unregulated growth and support of AI initiatives.
A panel discussion titled ‘Education, Culture, and the Environment in an AI-Driven Era’, moderated by Professor Michael Menchaca from the Department of Learning Design and Technology at The University of Hawai‘i at Mānoa, and comprising IAFOR President, Professor Jun Arima from The University of Tokyo, Japan; Professor Peter J. Mataira from Hawaii Pacific University, United States, and Dr LJ Rayphand, Dean of Outreach Education at Caroline College and Pastoral Institute in Chuuk State, Federated States of Micronesia, centred around contemporary effects of AI in the areas of education, culture, and the environment.
In his opening remarks, Professor Menchaca provided context for defining AI, explaining that AI is often presented as a tool for advancing knowledge and providing innovative solutions to complex problems. He defined AI simply as a tool capable of analysing vast amounts of information to identify patterns and ‘predict’ what might come next, whether a word, an image, a behaviour, an outcome, or even a trend. He provided the following example sentence for the audience to complete and predict like an AI system: ‘Even though we were behind in the game, our team still believed that we could ___’. Interestingly, the audience provided the same top three words predicted by AI to fill in the blank: win, succeed, and prevail. Professor Menchaca further described how an AI system, over time, might learn that humans had a preference for the word ‘win’ and predict that word statistically more often than others. He stated that AI continues to evolve and improve; however, such progress ‘comes with a price’, including immense environmental and challenging cultural impacts.
Regarding the environment, Professor Arima explained that AI can have both positive and negative effects on international climate issues and sustainability. On the positive side, he stated that AI can be used for predicting future electricity demand or load forecasting and matching supply to demand through grid balancing to maximise the use of renewable energy sources or renewable penetration to increase overall energy savings. AI can also be used for consistent, efficient production and predictive maintenance in the industry sector, as well as traffic optimisation in the transport sector. AI can also be a powerful tool for climate management, including early warnings for extreme weather, monitoring deforestation, and recognising biodiversity loss.
Despite the positive effect AI can have on reducing environmental degradation, what is commonly called ‘Green AI’ among industry experts, Professor Arima also talked about the paradox that is the increased electricity and water demands to adequately run and cool data centres due to the expansion of AI in all sectors, which requires the use of energy-consuming servers often powered by fossil-heavy grids, increasing CO2 emissions. According to Professor Arima, this paradox sees AI increasing demands on the environment while simultaneously providing potential solutions to those very demands. This raises a very important philosophical question around why we use AI: is it a solution to our problems, or are we experiencing a means-ends inversion, in which we create more problems for the sake of technological expansion? Professor Arima concluded that efficiency improvements lower costs and actually encourage and lead to higher consumption and digital divides where the concentration of AI capacity exists in only a few corporations or countries. According to him, AI’s impact is neither inherently good nor bad, and outcomes depend on how society, government, and industry steer its deployment.
Cultural bias and social injustice embedded within AI are additional risks to native and Indigenous communities in particular. Professor Mataira questioned whether algorithms can truly understand relationships and indigenous ways of knowing and being. Specifically, he doubted whether any AI system, especially those designed within a Western framework, could ethically apply such concepts as kinship, responsibility, or spiritual connection to land. He shared his own experience using AI tools that persistently imposed an alien worldview, perpetuating historical fantasies and misrepresenting Indigenous identities. He further explained that indigenous approaches are relational, holistic, and place-based, while AI is built on prediction, quantification, and decontextualised aggregated data. Key issues addressed included acknowledging cultural integrity, epistemological tension, technological limitation, and the risk of assimilation.
Regenerative trauma, algorithmic bias, and systemic harm to indigenous communities are also possible adverse outcomes of AI designed predominantly within Western frameworks. Professor Mataira compared AI to a guest in communities that may be welcome or unwelcome, invited or uninvited. AI systems trained on biased data risk automating and perpetuating cultural biases. He expressed a particular concern that AI tools used in educational settings could provide new methods for surveillance, disproportionate interventions based on cultural bias, and disruption of Indigenous families and cultural practices through a lack of cultural nuance in AI systems. Recolonisation, recurrence of historic trauma, systemic bias, and harm disguised as progress are some of the potentially indirect consequences.
Because the Western worldview tends to be unaware or ignore intercultural nuances outside of its own framework, it is important to ask ourselves whether the uncritical adoption of AI, controlled by non-Indigenous entities, is appropriate for speaking for Indigenous rights. ’Does embracing externally developed AI for “child well-being” risk a new form of digital colonisation of our children’s futures, minds, and relationships?’ Professor Mataira asked. What if, instead, Indigenous communities were to lead the design, governance, and ethical considerations for technology used with our children, ‘ensuring it strengthens rather than supplants our ways of knowing, doing, connecting, and being?’, he further questioned.
According to Professor Mataira, AI acts as a new way of recolonisation that strips Indigenous communities of their rights to self-determination. There are power dynamics at play that act through the acquisition and reproduction of data. Professor Mataira explained why training AI on data and controlling the narrative is crucial to understanding these power dynamics: data acts as the new land. According to him, ‘harvesting of data about our children, families, and communities… mirrors historical patterns of resource extraction and land dispossession’. He described AI systems as potentially similar to previous colonising efforts on native lands, leading to land acquisition, redistribution, and the extraction of resources. He wondered what true data sovereignty for Indigenous nations might look like and how to protect future generations.
In the same line of argument, Dr Rayphand also discussed the impact of AI on Micronesian culture and its oral traditions. In particular, he noted that Micronesia is historically less explored than other island regions, a concept described as ‘a hole in the doughnut’. Micronesians document, preserve, and pass down information through songs, chants, myths, and stories, with storytelling being a fundamental educational tool. He expressed concern that as technological systems such as AI increasingly evolve to replace the historical understanding and voice of indigenous peoples, stories and the knowledge they provide will not only be lost but replaced by knowledge with concerns about ‘validity, authenticity, and authority’. Dr Rayphand presented objects AI included as representations of Chuuk culture that were not actually native as an example of what happens when AI is not indexed appropriately.
Dr Rayphand stressed using AI to share authentic stories with Micronesians, looking to their own communities to decide the purpose and role of AI, most especially contributing their own data rather than simply retrieving information. In particular, he suggested that curriculum, syllabi, and policies need to be updated to address the ethical use of AI for Indigenous purposes. Dr Rayphand concluded that if Indigenous communities do not engage with and introduce AI into their schools and use it with their own voice, ‘somebody will do it for us’.
This discussion is important at a time when AI and other types of algorithms make us question the concept of knowledge, and its production and reproduction, including questions that we have seen repeatedly at recent IAFOR conferences, including: What place does non-written, oral, and tacit knowledge hold in today’s educational system? Especially in a world that has become more audiovisual than text-based, and in which students learn new knowledge through non-traditional educational tools such as TikTok or YouTube, how is Indigenous, non-written knowledge represented and taught?
Watch the full panel presentation at IICE/IICAH2026 in the video below.
Banner image: Igor Shalyminov, Unsplash
