Research and direct observation are essential for driving innovation in AI's creative process, according to Adewole Ampitan Adeyemi, an experienced digital product designer. In an interview with OPEYEMI BABALOLA, Adeyemi argues that the future of AI in product design depends less on speed and automation and more on how well products understand the realities of the people using them.
Challenging AI Assumptions with Real User Behavior
Adeyemi emphasized that real user behavior cannot be abstracted away or treated as secondary. While working at Interswitch Group on Quickteller Business and agency banking platforms, he encountered behaviors that data alone could not explain. For instance, fraud detection models flagged shared devices as suspicious, but in reality, one phone might serve an entire household or an agent processing transactions for dozens of customers. Similarly, credit models assumed structured monthly salary deposits, yet many users earned income daily through cash-based transactions.
These experiences revealed a gap between model assumptions and lived reality. Many AI systems assume users are individual, continuous, and consistently measurable, but real users are often collective, intermittent, adaptive, and only partially observable. Their behavior is shaped by context, infrastructure, culture, and economic realities that structured datasets do not capture well. UX research and direct observation remain essential even in AI-driven systems.
Designing for African Users vs. Global Data Patterns
Designing for African users, especially from ethnographic research in Lagos markets, taught Adeyemi that behavior is deeply shaped by environment. Global product assumptions come from highly connected environments with stable infrastructure, personal device ownership, and established digital trust. However, in Lagos markets, users share devices, rely on multiple SIM cards, check transactions with neighbors before trusting confirmations, and develop workarounds for unstable connectivity.
Trust is a major difference. Many global systems assume a successful transaction means the user moves on, but in African markets, trust is social and behavioral. African users are incredibly adaptive, improvising around infrastructure limitations, cost, connectivity, and familiarity. This experience changed Adeyemi's approach to design. He learned to design mobile-first as a constraint that sharpens every decision. Designing for sub-3G environments, low-end devices, and shared phones forces clarity and strips products down to what matters.
Ironically, those constraints often produce better products for everyone. Designing purely from global patterns can create products that scale technically while excluding people socially. African users expose weak assumptions hidden inside many modern products, forcing designers to confront questions about trust, resilience, accessibility, and adaptability that apply far beyond Africa.
The DesignFlow Kit: Solving a Critical Gap
Adeyemi developed DesignFlow Kit as an AI connector rather than a generator. Instead of just producing screens, it connects research insights, design decisions, and delivery workflows so that when one part changes, the others remain aligned. After nearly eight years leading design across Cyberspace, DOT, and Interswitch, he saw the same pattern: as design systems and AI tools improved at generating outputs, the real bottleneck became maintaining coherence as products evolved.
Research findings got lost between sprints. Design decisions made during workshops never fully made it into engineering handovers. Design system rules drifted out of sync. The issue was not the absence of artifacts but the breakdown of continuity between them. DesignFlow Kit creates traceability between research, design rationale, system standards, and delivery tasks, enabling teams to move quickly without losing context.
Adeyemi released it as open source so teams anywhere can adapt it to their operational realities, governance structures, and data constraints. Many closed AI design tools encode Silicon Valley assumptions about users, infrastructure, trust, and behavior that do not translate well across different environments. Existing AI tools help individual designers produce screens faster, but DesignFlow Kit helps entire design teams stay aligned while moving quickly.
Most Significant Contributions
Adeyemi's proudest contributions are in two areas. First, his work at Interswitch co-leading ideation and feature design for Quickteller Business and agency banking platforms, which serve millions of users across Nigeria. The USSD, QR payment, and withdrawal flows he designed were often used by underbanked individuals interacting with digital financial services for the first time. At that scale, improving navigation or reducing time-on-task becomes the difference between successful completion or abandonment. Clarity is access.
Second, DesignFlow Kit, the open-source AI-powered workflow automation product he built from the ground up. It brings together everything he learned from years of leading design across fintech and enterprise environments. It solves the problem of maintaining alignment between research insights, design decisions, engineering handovers, and governance as products evolve. Its openness allows anyone to adapt it to their own workflows rather than being forced into assumptions embedded in closed systems.
Experiences That Shaped His Thinking
Adeyemi's Physics degree at Covenant University taught him to ask what is actually happening underneath the interface. That habit of refusing to take any system's output at face value shaped his approach to user research and product thinking. He remains cautious with AI because the polish of an output can hide weak assumptions, missing context, or flawed reasoning.
His UX work at Cyberspace Limited and Interswitch Group, designing for underbanked users, reset his understanding of good design. Watching someone complete a transaction over USSD on a feature phone taught him that elegance is not the goal; access is. That insight has shaped almost every design decision he has made since.
He also realized that systems do not fail all at once. Breakdown happens gradually through small inconsistencies, weak assumptions, and disconnected decisions that compound over time. That pushed him toward systems thinking, not just designing screens but thinking deeply about how research, behavior, infrastructure, trust, and delivery connect. It led him to build DesignFlow Kit and care about creating products that remain coherent as they scale.
Principles for the Future of AI-Driven Product Design
Adeyemi believes AI should expand the agency of the user, not replace it. Most current AI design does the opposite: autocomplete narrows what you would have said, recommendations hide options, and smart defaults remove decisions users did not know they were giving up. At population scale, that compression of agency is a bigger long-term risk than hallucination or bias, and it is a design problem more than a technical one.
In practice, this means defaulting to augmentation rather than automation, surfacing uncertainty honestly instead of performing confidence, and keeping the user as the protagonist of their own decision. Products that take agency from users may win on engagement metrics in the short term, but over a generation they erode the very judgment they were supposed to support.
Common Mistakes Companies Make with AI
From a designer's perspective, the mistake is treating AI as a feature when it should be treated as a system. The pattern is the same across every sector: a model gets wrapped in an API, dropped into an existing product surface, and announced as an AI capability without asking what it does to the user's flow, trust, or sense of control. What is missing is everything around the model: data lineage, governance, human override, drift monitoring, feedback loops, and a UX layer that helps the user understand what the AI is doing and why.
The result is AI that looks impressive in a demo and underperforms in production, with no instrumentation to understand why it fails and a user experience that quietly erodes trust. The correction is treating AI governance and AI UX as first-class disciplines alongside engineering and security. Every previous wave from web to mobile to cloud went through this same arc from feature-first to systems thinking. AI is mid-arc right now, and the companies that move early will be the ones still trusted with AI in five years.
