The rise of artificial intelligence (AI) presents unprecedented challenges to the concept of citizenship in the twenty-first century. Digital citizenship, a fledging idea, is already in peril. Alongside it, potentially, democracy itself is at stake.
The notion of citizenship itself originated in ancient Greece and Rome — and evolved to the dominant idea of granting individuals certain rights within land-based nation-states. Today, digital technologies and the internet have expanded the “anthroposphere,” the realm of human interaction, to include physical and virtual spaces.
In essence, citizenship is progressing — and struggling to both keep existing rights in the face of new challenges and acquire new ones. In the eyes of its proponents, digital citizenship is the latest — and most progressive — iteration of this idea. But the shift to digital has not erased social inequalities — quite the opposite. Indeed, the rise of AI may exacerbate social inequalities related to digital citizenship.
Here’s why.
Our current digital geography mirrors offline social inequalities, making it difficult for historically marginalized groups to create content and access reliable information. Those with limited proficiency with technological tools find their access to online public services restricted and their abilities to discern between real and false content online hampered, which affects their capacity to exercise citizenship. A timely example is the torrent of false images emerging in the 2024 US presidential race. It’s telling that the various interpretations of digital citizenship all advocate for individuals’ ability to fully participate in digital society as a premise. But these challenges in the current digital public sphere are obstacles to informed, fact-based participation by all.
While our digital societies grapple with these challenges, AI introduces further complexities. What will happen when it is not only individuals but also machine entities asserting citizenship rights online? How will regulators respond if AI-powered corporations come to wield influence surpassing that of entire cities or nations? As our previous studies have shown, technological developments often have unintended consequences.
Indeed, digital citizenship is at risk of perishing in the cradle, due to the still largely unregulated applications of AI. Moreover, digital citizenship can be engineered, exploited or even rendered harmful through AI-driven online manipulation. Citizenship — like democracy — does not necessarily evolve in a linear, progressive fashion; it experiences setbacks, obstacles and attacks. But the trajectory can still change for the better.
In our view, the widespread adaptations of AI present three new challenges to this notion of citizenship.
First, digital citizenship is about participation and inclusion in the digital society, thereby ensuring that individuals can benefit from it. To navigate and thrive in the digital world, individuals need to acquire certain competencies, which we usually call digital literacy. However, AI systems are likely to exacerbate the existing digital divides among people that are related to their competencies in the use of technologies, so that those who are less skilled get left further behind. Related to these increased pressures for digital literacy, the automation facilitated by AI is likely to lead to job displacement — affecting different parts of society differently. Researchers at the University of Oxford estimate that anywhere from nine to 32 percent of jobs could be at risk of being automated in future, leaving some without income and intensifying socio-economic divides.
Second, the emergence of generative AI chatbots that are capable of mimicking human behaviour raises concerns about the legitimacy of online interactions. These “AI personas” will be able to imitate genuine citizens and, with that, perhaps acquire political influence. Deepfake video and audio recordings already pose a threat to the integrity of online information — undermining the digital public sphere. These AI-created pollutions of the online environment can skew the online conversation further and make digital worlds the very worst place for citizens to inform themselves — obstructing democratic deliberation, and further impeding the development of literate digital citizenship.
Third, the dominance of large social media platforms has concentrated significant power in the hands of a few. These platforms leverage data, algorithms and potentially AI to influence not only online spaces but also physical ones through policy and business models. And, they are the best positioned to train new AI with all their users’ accumulated data. This concentration of power (data) could be further amplified by AI, potentially before effective regulations or counterbalances are established. Some experts, such as The Future of Life Institute and researchers from the National University of Singapore, suggest that large tech companies are already trying to amass power before regulation comes. Additionally, as AI and hardware become increasingly intertwined, tech giants such as Microsoft may leverage this to solidify their dominance in hardware and software markets.
In sum, AI is poised to make challenges to democracy more potent — in this case, by interfering with the developments of digital citizenship worldwide. This risk calls for urgent deliberations and reassessments of digital citizenship practices — and includes the need for innovative citizen technologies that factor in emergent methods of voter manipulation and deal with new social and digital inequalities that will affect the enjoyment of rights.
Civil society and the global research community should consider what practices of digital citizenship work to uphold democratic principles and ensure the fulfillment of rights within this rapidly evolving landscape. Addressing new challenges demands more robust digital citizenship practices. These practices should not only counter manipulation but also promote equitable access to the benefits of these new technologies from a human rights-based perspective. Additionally, policy makers should be guided by principles of digital citizenship when adapting legislation to meet the new challenges posed by AI.
Ideally, digital citizenship involves practices aimed at upholding the achievements and rights established in previous eras, while leveraging the opportunities presented by digital society and the democratic potential of new technologies.
Reassessments along these lines should be based on current research on how the concept can be understood from a rights perspective.
We argue for digital citizenship practices that respect human rights and democratic values, and for interactions in these new spaces that address, rather than amplify, existing societal inequalities. From their early design, through their implementation, to the evaluation of their impacts, we need to consider all the ways in which emerging technologies affect citizenship.
