By Omoruyi ‘Uyilaw’ Edoigiawerie, Esq
Two years ago, a business owner proudly walked me through what he described as the engine of his company’s growth. It was not an exceptional sales team, nor an elaborate distribution network, nor even a revolutionary product innovation. His confidence rested on something less visible but immensely powerful: an algorithm. More accurately, access to one.
His business had mastered digital customer acquisition with remarkable precision. Demand came predictably, engagement was strong, and growth felt almost engineered.
For a time, it looked like entrepreneurial brilliance. Then the algorithm changed. Quietly. No consultation, no negotiation, no regard for the assumptions upon which an entire business model had been built.
Reach collapsed, acquisition costs surged, and predictability vanished. The business remained legally his, of course, but a harder truth had emerged: a critical part of its prosperity had always been controlled elsewhere.
That story returns to mind in today’s race to build artificial intelligence-powered businesses. Across start-up ecosystems, artificial intelligence has become the language of relevance.
Founders invoke it in pitch decks, product teams weave it into demonstrations, and investors increasingly expect to hear some articulation of an AI strategy. Yet beneath the excitement lies a question deserving more serious reflection than it currently receives: who owns the algorithm?
At first glance, it appears to be a technical question, the sort best left to engineers. It is not. It is fundamentally a business question, and increasingly a governance one. For emerging economies like ours, it may even be a strategic question.
A significant number of founders building AI-enabled products today rely on intelligence systems they neither own nor control. They integrate powerful external models into customer-facing products and construct commercial offerings around their capabilities.
There is nothing inherently wrong with this. Businesses have always relied on infrastructure they did not personally build. Airlines do not refine jet fuel. Banks do not manufacture telecommunications equipment. Retailers do not construct highways. Dependency, in itself, is not unusual. What matters is understanding the nature of the dependency.
A founder may build an intelligent legal assistant, another may automate customer service, while others deploy machine intelligence for lending decisions, logistics optimisation, content production, or compliance monitoring.
The products may be commercially viable and genuinely useful. Revenue may grow. Customers may even become deeply engaged. But the question remains: where does ownership truly reside?
The interface may belong to the founder. The customer relationships may belong to the founder. The workflows may belong to the founder. But the intelligence powering the experience may belong entirely to someone else.
This matters because ownership determines influence. The party that controls the underlying intelligence layer controls pricing, access, technical limitations, and strategic direction. In some cases, it may also become a competitor as easily as a supplier. That is not cynicism; it is simply market behaviour.
A founder whose differentiation rests largely on access to intelligence controlled elsewhere must ask uncomfortable but necessary questions.
What happens when pricing changes? What happens if usage terms evolve? What if the provider imposes restrictions? What if reliability becomes uncertain? More pointedly, what if the provider decides your market segment is attractive enough to enter directly?
Technology history offers familiar warnings. Businesses have built around social media distribution they did not own, search visibility they could not control, and advertising ecosystems whose economics shifted without notice. Artificial intelligence presents a more sophisticated version of the same structural risk.
This is not an argument against building with AI. On the contrary, artificial intelligence offers extraordinary opportunities to unlock efficiency, create new products, and solve longstanding problems.
Ignoring it would be strategically short-sighted. But there is a difference between leveraging powerful technology and misunderstanding one’s relationship to it. Access should never be confused with ownership.
For African founders, this conversation carries added urgency. Much of the digital infrastructure powering innovation across the continent is already under external control.
Cloud services, enterprise software, advertising systems, payment rails, and developer infrastructure are often governed elsewhere.
Artificial intelligence threatens to deepen this dependency. The issue is not whether global collaboration is beneficial; it clearly is. The real issue is whether local ecosystems are building enough proprietary advantage within that collaboration.
If the intelligence powering the next generation of African businesses is externally owned, externally priced, externally governed, and externally prioritised, where will durable economic power ultimately reside?
This is not ideological rhetoric. It is a commercial reality. Businesses dependent on infrastructure that they cannot influence are structurally exposed. Decisions made in distant boardrooms can materially affect local business economics.
There is also a legal dimension that many founders have yet to confront with sufficient seriousness. Artificial intelligence raises practical questions about data rights, liability, transparency, regulatory compliance, and commercial risk.
Who owns generated outputs? What liabilities arise when automated decisions fail? How should confidential customer data be treated where third-party systems are involved? For founders in finance, healthcare, legal technology, education, and other regulated sectors, these are not abstract matters.
Institutional customers will increasingly ask these questions. Regulators certainly will. Sophisticated investors already are.
The market is evolving. The first phase of AI enthusiasm rewarded simple proximity to artificial intelligence. Having an AI narrative often seemed sufficient. That phase will pass. Markets eventually become less impressed by novelty and more interested in defensibility. What is proprietary? What survives commoditisation? What remains valuable if the underlying intelligence becomes widely accessible?
Not every successful AI-enabled company will need to own foundational models. That would be an overly simplistic conclusion. Many will succeed through trust, execution, specialised data, domain expertise, workflow integration, and regulatory competence. Those are real assets. But founders must be honest about where their true value lies.
The most important question is no longer whether your business uses artificial intelligence. Increasingly, that is the wrong question. The better one is this: what do you own that cannot be switched off, reprised, replicated, or withdrawn by someone else?
That is where enduring value resides.
*Omoruyi “Uyilaw” Edoigiawerie is a leading start-up lawyer and policy advisor working at the intersection of law, technology, and equity in emerging markets. He is the Founder and Chief Servant at EandC Legal, a full-service law firm offering bespoke legal services to start-ups, established businesses, and upscale private clients in Nigeria. The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances. To get in touch, please email: hello@uyilaw.com.
