The AI Development Fallacy: Why Technology Alone Won't Transform the Global South

The recent essay by Dario Amadei “The Machines of Loving Grade” proposes rosy projections about using AI to rapidly boost economic growth and development in the poorest parts of the world fail to grapple with the stark realities in those societies. The author paints a picture of AI-powered epidemiology and economic planning sweeping away entrenched barriers to progress, but the track record of technological leapfrogging in the developing world tells a very different story.

Take the example of mobile internet technology. Over the past two decades, we've seen unprecedented adoption of smartphones and internet access even in the remotest corners of the global south. Yet this "mobile revolution" has had vanishingly little impact on productivity, economic growth, or standards of living. Researchers have struggled to find evidence that these transformative information technologies have translated into meaningful development outcomes.

The reasons are manifold. Lack of complementary infrastructure, skills gaps, and institutional weaknesses have prevented communities from fully harnessing the potential of mobile technology. Connectivity alone does not automatically create economic opportunity or improve quality of life. The obstacles are as much social and political as they are technical. The “education crisis” which means the majority of the population can not use these technologies to improve their productivity. Last but not least, the massive inequality and poverty which is not a problem of lack of resources but an issue of access to power and representation.

The essay blithely asserts that AI-powered economic planning and policymaking can drive 20% annual GDP growth in the developing world. But this flies in the face of decades of failed attempts at top-down, technocratic development schemes. From the Green Revolution to Structural Adjustment Programs, grand plans hatched by development economists have consistently fallen short when confronted with the complexities of on-the-ground realities.

The author argues that AI could help address issues like corruption and weak governance that have undermined past development efforts. Yet these entrenched political and institutional challenges are precisely the kind that are least susceptible to technological solutions. Powerful AI may give authoritarian rulers even more tools to surveil, manipulate and repress their populations. Without fundamental reforms to systems of accountability and the rule of law, AI is just as likely to entrench the status quo as it is to catalyze positive change.

Moreover, the essay overlooks the was in which advanced automation driven by AI could actually exacerbate inequality and social instability in the developing world. As AI-powered technologies displace large segments of the workforce, the potential for social unrest and political backlash is very real. Transitioning agrarian economies to knowledge-based, AI-driven development is a monumental challenge that will require painstaking, context-specific approaches - not the kind of one-size-fits-all techno-solutionism on display here.

In the end, the essay's vision of AI-fueled development reads as little more than Silicon Valley utopianism writ large. It fails to grapple with the deep-rooted historical, political, and institutional barriers that have stymied progress in most societies in the global south, even in the face of prior technological revolutions. Unless these underlying challenges are directly confronted, the prospect of AI driving rapid, equitable development remains a pipe dream, no matter how dazzling the technological capabilities on display.

Sources: Amodei, Dario. “Dario Amodei — Machines of Loving Grace.” Darioamodei.com, 11 Oct. 2024, darioamodei.com/machines-of-loving-grace#basic-assumptions-and-framework. Accessed 7 Nov. 2024.

‌