This is as an entry for the Essays on longtermism Competition.
When I first encountered longtermism through What We Owe the Future and later in discussions within the EA community in Nigeria, what struck me most was not the distant timelines or abstract population ethics debates. It was the simple but profound claim that our actions today can meaningfully shape the trajectory of generations we will never meet. This insight felt especially resonant coming from a context where institutions often struggle to plan even a decade ahead, let alone a century.
The recently published Essays on Longtermism: Present Action for the Distant Future captures this challenge in a rigorous academic way. One chapter in particular — Jacob Barrett and Andreas T. Schmidt’s “Longtermist Political Philosophy: An Agenda for Future Research” asks us to consider what kinds of political and institutional structures are best suited to safeguard the long-term future. Their questions about legitimacy, representation, and feasibility are pressing, but they are also, in some ways, incomplete.
While much of the discussion of longtermist priorities is rooted in American and European institutions, there is a pressing need to expand this analysis to the Global South. Transformative technologies such as artificial intelligence (AI) will not only affect powerful states but will reshape societies everywhere. For longtermism to succeed in practice, its institutional grounding must be global, inclusive, and capable of addressing asymmetries of power in technology governance.
This essay responds to the chapter “Longtermist Political Philosophy: An Agenda for Future Research” by Jacob Barrett and Andreas T. Schmidt, extending their framework into the context of AI governance in the Global South. I argue that building institutional readiness for AI in low and middle-income countries is an essential component of longtermist action, one that has been comparatively underexplored but holds significant implications for humanity’s long-term trajectory.
Barrett and Schmidt note that longtermism demands serious engagement with institutional questions: what kinds of political structures are best suited to safeguard the long-term future? Their analysis highlights issues such as legitimacy, intergenerational representation, and feasibility. Applied to AI governance, these questions become even more urgent, as the rapid development of advanced AI could dramatically reshape global power structures and human welfare.
From a longtermist perspective, the stakes are unusually high. AI carries both extraordinary opportunities (scientific breakthroughs, economic growth, enhanced decision-making) and catastrophic risks (misalignment, misuse, disempowerment of human agency). The path we take now will reverberate through centuries, perhaps millennia. Thus, designing inclusive governance frameworks today is not merely desirable but morally required.
Much of the discourse on AI safety and governance takes place in advanced economies such as the U.S., the EU, and increasingly China. Yet the Global South, where the majority of humanity resides, faces unique vulnerabilities:
Ignoring the Global South in AI governance frameworks risks entrenching a two-tier future: one where technological benefits accrue to a minority, while risks disproportionately affect the majority. Longtermism, properly understood, cannot allow this outcome.
How might longtermism inform practical action in building institutional readiness in the Global South? Here are three pathways I believe longtermists should take seriously:
Embedding Longtermist Principles in Regional Institutions
Regional organizations such as the African Union, ECOWAS, and ASEAN can serve as testing grounds for embedding intergenerational ethics into policy frameworks. For instance, chartering advisory councils dedicated to long-term technological risks could institutionalize foresight capacities in a way aligned with longtermist values.
One might object: aren’t these organizations already overburdened with urgent crises? Yes, but this is where longtermism’s value becomes clear. Many of the near-term crises like climate adaptation, food security, technological inequality, are exactly the ones that longtermist frameworks can help navigate, because they demand thinking beyond election cycles or donor funding rounds.
Inclusive Global Governance
Global initiatives on AI standards, safety, and coordination must not remain exclusive clubs of technologically dominant nations. Longtermists can advocate for inclusive decision-making structures, ensuring that states representing the majority of humanity have a voice in shaping the trajectory of transformative AI.
It is not enough for powerful states to invite token representation. We need genuine participation, where perspectives from Lagos, Nairobi, or Dhaka shape the norms and standards that will govern transformative technologies. From a forecasting perspective, diverse viewpoints improve our models of the future. Narrow epistemic circles miss crucial risks.
Some critics of longtermism argue that its focus on the distant future risks neglecting present injustices. Of course, some might argue that focusing on the Global South dilutes scarce longtermist resources. Shouldn’t we prioritize where the cutting-edge labs are, Silicon Valley, London, Shenzhen?
This objection has force, but I think it misunderstands the point. Longtermism is not only about mitigating the risks of frontier technologies; it is also about shaping the global distribution of resilience. If a misaligned AI emerges tomorrow, yes, the labs matter most. But if AI transforms society gradually, which many experts think is more likely, then the question of who benefits, who loses, and who governs becomes just as existential.
This is precisely the type of “dual-use” action that strengthens longtermism’s practical appeal.
Reading Essays on Longtermism made me realize how much more work remains to connect the lofty philosophical arguments to the messy realities of global politics. Barrett and Schmidt ask: what political structures can represent the interests of future generations? My answer, building on their framework, is that those structures must not be confined to wealthy democracies.
The distant future will not be decided in Oxford seminar rooms or Palo Alto boardrooms alone. It will be decided in Abuja, Dhaka, São Paulo, and Nairobi, in the choices that billions of people make, and in the institutions that guide them. If longtermism is to succeed, it must expand its scope to include these voices, these contexts, and these futures.
For me, as someone working on community building in Nigeria and engaging with questions of AI safety, this is not an abstract claim. We already see glimpses of both promise and peril in how technologies are deployed here. The question is whether we will have the foresight, and the courage, to prepare our institutions now.
Longtermism asks us to take responsibility not just for the present, but for the arc of history that stretches beyond us. And if that responsibility is real, then the Global South must be part of the story, not as an afterthought, but as a cornerstone of humanity’s shared future.
Longtermism, as framed in the Effective Altruism community, deeply worries me because it feels too confined to that intellectual circle and risks being detached from reality. If it is to have genuine value, it must be applicable beyond this space—not just as a lofty ideal, but as something that speaks to and engages with the real world.
My concern is that longtermism, in its current practice, often seems to prioritize the good of the distant future in isolation from the urgent needs of the present. This separation fills me with anxiety. The future cannot be secured while ignoring the present, because the present is the launchpad for whatever comes ahead. To shut our eyes and ears to today’s pressing calls is to weaken the very foundation on which the future depends.
Longtermism must not work in isolation. It must engage the present, test its ideas in communities, and prove its relevance here and now before projecting them forward. Otherwise, it risks becoming a paralyzing concept —frozen in speculation while real people struggle. How can we dash into the future while leaving today’s crises unresolved?
This concern weighs even more heavily when I think of the Global South. These regions already carry enormous burdens—poverty, unemployment, food insecurity, fragile infrastructure. For them, an abstract future-focused framework that sidelines the present may not just be irrelevant but counterproductive. Unless longtermism learns to work in two time zones at once—the present and the future—its interventions may harm rather than help.
Thank you for raising this
I think your concern strikes at something many of us within and around the longtermist community have been reflecting on. I share the worry that longtermism can sound detached or even abstract when it’s presented purely as a philosophical ideal rather than as something that must earn its relevance through present-day impact.
But I don’t think that the tension between “present” and “future” is an unavoidable flaw. In fact, part of my motivation for writing this essay was to show that the two are deeply connected, especially in contexts like the Global South, where long-term risks and present-day vulnerabilities are intertwined.
For example, when we talk about building institutional readiness for AI, we’re not talking about ignoring current crises. We’re talking about strengthening the very institutions that can address them, improving education, governance, foresight, and inclusion. These are present actions that both reduce near-term harms and make our societies more resilient to long-term risks.
In that sense, I see longtermism not as an escape from the present, but as an invitation to act more wisely within it. The idea is not to postpone compassion or justice for the sake of distant futures, but to ensure that today’s solutions don’t mortgage tomorrow’s possibilities.
I completely agree that longtermism must prove its worth “in two time zones at once.” For those of us in the Global South, that dual engagement isn’t optional, it’s survival. We can’t talk about safeguarding the year 2500 if we can’t feed, educate, or protect people in 2025. But neither can we afford to build only for 2025 if the systems we create are brittle, exclusionary, or unprepared for transformative change.