Lately, I’ve been binge-reading on artificial intelligence (AI) and delving even deeper into fiction and non-fiction by African authors, primarily those reflecting on Africa in the post-colonial period. Interestingly, these seemingly disparate paths have converged, leading me to reflect on the potential impacts – both positive and negative – of AI on Africa, and what we as Africans should be doing to shape our own AI destiny.
Africa knows all too well what can happen when its future is determined by a few people sequestered in closed-door “by invitation only” sessions in far off well-appointed places – the Berlin Conference on Africa of 1884-1885, comes to mind. Not a single African was present, but that didn’t stop the organizers and attendees from dictating the entire continent’s (a landmass three times larger than Europe) fate from then until now, including where national boundaries in Africa would be drawn, which people belonged where regardless of ethnicity, what official language would be spoken by the people, which European power would rule them, and so on and so forth.
The debate on AI unfolds against the backdrop of a global geopolitical landscape undergoing seismic shifts and redefining the Post World War II world order. As Africa navigates and negotiates its place in it, it is crucial that we learn from the mistakes of the past and ensure that the voices of all those who will be impacted by this powerful and transformative technology enhancer are heard and considered in shaping its future. To help us navigate this complex terrain, we need guideposts that provide insight and context.
One such guidepost is Dr. Fei-Fei Li’s memoir, “The Worlds I See.” In addition to being a great read, it provides a unique lens through which to view the interplay between AI and Africa.
In “The Worlds I See,” Li provides an essential and clear account of the origins, evolution, and current state of AI – the good, the bad, and the ugly. She also raises the question of who will win, the good guys or the bad guys, especially when she highlights who has the power in AI, both the funding and the data on which most AI models are being built:
“What worried me, though, was that every conversation I had on the matter, without a single exception, ended with the same plea: that the research they found most interesting wouldn’t be possible outside of a privately run lab. Even at a place like Stanford, the budgets just weren’t big enough. Often, in fact, they weren’t even close. Corporate research wasn’t just the more lucrative option; it was, increasingly, the only option.”
– Dr. Fei-Fei Li, “The Worlds I See”
Li’s American immigrant story serves as the perfect metaphor for the outsider in the AI world, even after reaching the highest echelons of the field. She illustrates what changes and, more alarmingly, what doesn’t change when an immigrant, particularly one who is not a white male, ascends to the top in areas like finance and tech, two areas that are doing the most to shape the AI world (and consequently, the real world) we live in. Despite her remarkable achievements and position at the forefront of AI research, she is still an outsider in a field dominated by a narrow demographic.
In a particularly poignant moment early in the book Li tells us about a violent and disturbing incident involving a classmate in her English as a Second Language (ESL) class of four Asian students. One of them was assaulted by their American classmates and was left with a concussion, a broken nose, and severe emotional trauma. The incident also traumatized Li and the other two ESL students who witnessed the assault. Even worse, because they were all “outsiders,” they were overwhelmed and felt powerless to intervene, or even report the incident to the school principal, fearing retaliation and that they might jeopardize their already precarious immigrant status. Nevertheless, this incident, coming just when she was beginning to feel that she “belonged,” was a rude awakening. It influenced her world views and, many years later, her work in AI, as well as her drive to make AI more inclusive and representative of all humanity.
Li’s story underscores the need for diversity in AI, especially as it becomes increasingly intertwined with every facet of our waking (and even sleeping) lives.
Here are some of my takeaways from her book:
- Inclusivity and Ethics Are Non-Negotiable: Committing to inclusivity and ethics in AI development is fundamental to creating technology that is equitable and benefits a diverse population.
- Diversity in Data Sets is Crucial: Ensuring diverse datasets is essential to avoid biased AI outcomes that may not adequately represent all societal and national groups.
- Testing Algorithms is Essential: Proper validation of algorithms is crucial to minimize risks and inaccuracies in AI applications, which could lead to potential harm.
- Transparency in Decision-Making Enhances Trust: Clear and open decision-making processes are vital for building trust and accountability in AI technologies, facilitating easier identification and resolution of issues.
- Accountability Must Be Prioritized: Establishing accountability mechanisms in AI systems is critical for sustaining user trust and promoting responsible technology use.
- Addressing AI Risks Through Regulation and Policy: As AI development shifts from academic research to corporate-driven innovation, governments must implement regulations and policies that ensure transparency, fairness, and accountability at the same pace as business innovation. Failure to do so could lead to a lack of oversight, increased inequality, and the potential for digital authoritarianism. Policymakers must work closely with businesses, academics, and other stakeholders to create a regulatory framework that promotes the responsible development and deployment of AI technologies.
“The Worlds I See” is a well-timed reflection on the intersection of technology, humanity, and business, and what it is like for someone from the outside who, through merit, persistence, and determination, made it into the innermost sanctum of AI – and later returns to her earlier world to share the secrets – like a reverse native informant.
Li’s memoir makes a powerful case for a technological future shaped by all those who will ultimately be impacted by it. It is an approach that is more likely to result in a positive outcome for all of us – where the good guys win and where AI and related technologies and applications reflect the worlds “we all” see, no matter where we come from. The UN projects that by 2050, Africa’s population will reach close to 2.5 billion, accounting for about 25% of the world’s population. This figure is expected to rise to nearly 40% by the end of the century. AI needs to reflect this, and it needs to start now.
As I thought about the convergence of my readings on AI and post-colonial African literature, I realized that Li’s memoir reminded me of another book I recently finished, Cheikh Hamidou Kane’s 1961 “Ambiguous Adventure,” which is also an immigrant story in which the protagonist, Samba Diallo, grapples with the challenges of navigating his rapidly changing worlds – post colonial Senegal and as a foreign student in France. In “Ambiguous Adventure,” Kane’s message is that people from all nations, particularly those considered “under-developed,” should actively participate in shaping the future of the emerging global landscape.
“In the city which is being born such should be our work—all of us, Hindus, Chinese, South Americans, Negroes, Arabs, all of us, awkward and pitiful, we the under-developed, who feel ourselves to be clumsy in a world of perfect mechanical adjustment.”
– Cheikh Hamidou Kane, “Ambiguous Adventure”
He goes even further, as he contemplates life in a world in which “we the under-developed” have played no part in shaping and for which we are therefore always ill at ease, unwelcome, and second-class, experiencing the kind of pain and marginalization that Li and her ESL classmates endured in the wake of the classroom assault, a trauma that cuts so deep that the scars never heal.
As Kane’s protagonist prays, “God in Whom I believe, if we are not to succeed, let the Apocalypse come! Take away from us that liberty of which we shall not have known how to make use. May the arbitrary power of Thy will throw out of order the stable course of our laws.…”
Fei-Fei Li’s “The Worlds I See” is a clarion call-to-action for greater inclusivity, diversity, and ethical considerations in the development of AI. As Africa and the world at large grapple with the transformative potential of this powerful new technology, it’s crucial that we heed the lessons from Li’s memoir and the wisdom from Kane’s “Ambiguous Adventure,” learn from the past, and work together toward an AI future that benefits and reflects the worlds we all see.
This article originally appeared on LinkedIn.
