Nearpays winning AI for Good's Innovation Factory is the kind of African AI story the global stage rarely rewards — a Lagos company judged on deployment, not decks.
At the Grand Finale of the ITU's AI for Good Global Summit in Geneva on 9 July 2026, the Nigerian fintech founded by chief executive Victor Daniyan took the year-long competition's crown, as the ITU announced. Nearpays builds SoftPOS technology that turns an ordinary smartphone into a payment-acceptance device for small merchants — no terminal hardware, no upfront cost. Guest judge will.i.am matched the $20,000 prize from his own pocket, taking Nearpays to $40,000, and health-AI finalists Doto Health and Nemocare each left with $20,000.
Why a Payments App Won an AI Summit
The judgement criteria are the story. Innovation Factory runs across a year, testing companies on traction, deployment and measurable good — not stagecraft. SoftPOS matters in exactly those terms: across Nigeria and much of the continent, card and contactless acceptance fails at the last mile because terminals cost money merchants do not have. Software that turns the phone a trader already owns into the terminal removes the hardware tax on participation. Research on financial inclusion shows acceptance infrastructure, not consumer wallets, is the binding constraint — Nearpays aims at precisely that constraint.
The win also lands against a continental backdrop this site tracks closely: Nigeria just became Africa's highest-ranked country on the Global Index on Responsible AI, and our reporting on the data-centre sovereignty question and the readiness gap shows the pattern — African AI ambition is real, and the infrastructure deficit is real, and both truths hold at once. A Geneva stage rewarding a Lagos deployment is a data point on the right side of the ledger.
The Quieter Announcement: Identity for AI Agents
The summit's closing day carried the announcement with the longer half-life. ITU standardisation chief Seizo Onoe's takeaways from the summit confirm a new ITU focus group on making digital identity and trust frameworks ready for AI agents, alongside an AI for Good Lab to help developing nations and an AI Standards Exchange Database aggregating standards across bodies. The agent-identity work answers a question arriving faster than most institutions admit: when software acts — pays, books, negotiates, files — on a human's behalf, who is the actor, and how does the counterparty verify the authority?
The two announcements are one story wearing two outfits. A merchant accepting payment through a phone needs to trust the payer; a customer whose AI agent pays needs the merchant to trust the agent. According to the ITU's framing, identity and trust frameworks are the rails both run on. The countries that helped write the payment standards of the card era collected the fees of the card era — and the standards work starting now decides who collects in the agent era.
Standards are the quietest form of power. Whoever writes the rules for agent identity writes the terms on which a billion people's software gets to act.
The Seat at the Table
The reason both stories belong together on this site's Africa beat is participation. We wrote about the UN's global AI commission when Geneva last convened on governance; the standards machinery now spinning up at the ITU is where the abstractions become field lengths and protocol handshakes. What I call Emergent Intelligence (EI) — the dignity-first frame for what the industry calls AI — measures these processes with one question: are the people the systems will act upon present where the systems are being specified? A Lagos company on the winner's podium is presence. African delegations inside the agent-identity focus group would be power. The first happened on 9 July; the second is the assignment.
The evidence from the summit gives the continent a concrete agenda: join the focus group early, seed the Standards Exchange Database with African regulatory experience, and treat the AI for Good Lab as an instrument rather than a gift. Data from the mobile-money era demonstrates what happens when Africa arrives early to a standard — M-Pesa's rails shaped global thinking on financial inclusion for two decades. Agent identity is the same opportunity, one era later.
💡Key facts: Nearpays (Lagos; CEO Victor Daniyan) won the AI for Good Innovation Factory 2026 on 9 July in Geneva — $40,000 after will.i.am matched the $20,000 prize; Doto Health and Nemocare took $20,000 each. Summit close (10 July): ITU launched a focus group on digital identity and trust frameworks for AI agents, an AI for Good Lab for developing nations, and an AI Standards Exchange Database.
Frequently Asked Questions
These are the questions readers have been asking since the Geneva finale. Short answers follow, drawn from the ITU's announcements and the financial-inclusion research.
What is Nearpays and what did it win?
In short, Nearpays is a Lagos-based fintech whose SoftPOS software turns ordinary smartphones into payment-acceptance devices for small merchants. The answer, simply put, is that Nearpays won the ITU's year-long AI for Good Innovation Factory 2026, taking $40,000 after guest judge will.i.am doubled the prize. The key is what the jury rewarded: deployment among real merchants, not a demo.
How does SoftPOS technology work?
SoftPOS uses the NFC radio already inside a standard smartphone to accept contactless payments, replacing dedicated terminal hardware with software. According to inclusion research, the acceptance side — merchants able to take digital payment — is the binding constraint in African markets, and removing the terminal cost removes the constraint's sharpest edge.
Why is the ITU working on identity for AI agents?
Because agents increasingly act on humans' behalf — paying, booking, filing — and analysis shows existing digital-identity frameworks assume a human at the keyboard. The answer is a focus group launched at the summit's close to make identity and trust frameworks agent-ready: who the agent represents, what authority the agent carries, and how counterparties verify both.
Who benefits from the new AI for Good Lab?
Developing nations, by design — the Lab exists to help them adopt AI solutions and participate in standards work, alongside the new AI Standards Exchange Database that aggregates standards across bodies. In other words, the machinery for showing up is being built; the showing up remains sovereign work.
What are the implications for Africa's AI decade?
Evidence from the week demonstrates the pairing that matters: African companies can win on global deployment criteria, and the standards governing the next era are being written now. Data from the mobile-money era shows early standard-setting compounds for decades — M-Pesa shaped global financial-inclusion thinking. The research agenda for African institutions is participation in the agent-identity work while the concrete is wet.
Sources
