Prove your AI systems are governed, before the law requires it.
Afrispan AI Assurance is an independent AI deployment assurance practice for Nigerian and West African enterprises. We verify what your AI systems actually do, with evidence, not vendor assertion, the same discipline financial audit has applied to accounting for a century, now applied to AI.
Nigeria's AI obligations are coming. The enforcement body isn't ready yet.
Nigeria published its National Artificial Intelligence Strategy in September 2025, a five-year vision to 2029. A National Digital Economy and E-Governance Bill, expected to pass in 2026, would give NITDA formal, risk-based regulatory authority over AI, including mandatory licensing and annual impact assessments for high-risk systems in finance, public administration, and automated decision-making. Penalties reach 10 million naira or 2 percent of annual gross revenue, whichever is higher.
As of March 2026, Nigeria's own independent AI Governance Regulatory Body, the strategy's own capstone institution, has not been constituted, and NITDA's Code of Practice for AI remains unfinalized. Enterprises know binding obligations are coming, but there is no operational government channel yet to prepare against them.
Private-sector assurance, built to the rigor the eventual law will demand, is the practical bridge, not a workaround. The mandatory annual impact assessment described in the coming bill is functionally the same deliverable as Afrispan's own Conformity Report and FRIA tooling, already built and demonstrable today.
Read the full regulatory case →Six ways we help you prove your AI systems are governed
Every service line is grounded in a real, working technical capability, not a description written ahead of the capability existing.
Regulatory Conformity Assessment
Scored against EU AI Act, NIST AI RMF, NIST AI 600-1, and ISO/IEC 42001 simultaneously. Untested obligations marked, never omitted.
ProvenFundamental Rights and Impact Assessment
Aligned directly with Nigeria's coming annual impact assessment requirement, with every finding tagged by its evidentiary basis.
ProvenCross-Border Deployment Framework
An honest check of whether your compliance evidence actually transfers between Nigeria, Ghana, and wider ECOWAS.
ProvenEvaluation and Red-Team Engagement
Cross-model judging, adversarial testing mapped to OWASP LLM and Agentic Top 10 and MITRE ATLAS, and drift monitoring.
BuildingGovernance Orchestration and Human Oversight
Live infrastructure, not policy documents: automated routing, an independent kill-switch, and a named-person resume mechanism.
ProvenComplex Workflow Automation, Governed by Design
AI-powered business automation with the same assurance discipline built in from day one, not retrofitted after deployment.
ProvenBuilt and demonstrated, not just proposed
Afrispan's technical capability is not aspirational. It spans three connected open-source projects, publicly demonstrable on GitHub, so a technical stakeholder can verify the work directly rather than take a claim on faith.
Nine governance phases
Real Gemini API execution throughout. A working kill-switch with immutable, hash-chained audit logging. The FRIA, Conformity Report, and Cross-Border Framework generators.
ProvenProduction evaluation architecture
Thirteen notebooks, cross-model judging, red-teaming mapped to OWASP and MITRE ATLAS, a verified CI/CD pipeline on GitHub Actions.
BuildingLive governance orchestration
Three-queue routing, a shared kill-switch gate, pre-committed evaluation verdicts, and a real-time dashboard, built and verified on real hardware.
ProvenBuilt by someone who has done both sides of this work
Steve brings over a decade of technology, risk, and governance experience built across Nigeria, the United Kingdom, and global technology platforms. He progressed into a dedicated Technology Risk and Data Governance Analyst role in a regulated environment, where he authored data governance policy that achieved full regulatory audit readiness and delivered bi-monthly C-suite risk reporting across a two-hundred-person organisation.
That governance grounding was built across borders, not one market. In the United Kingdom, he worked as a Cybersecurity and Data Risk Analyst for Althaus Digital, and he brings direct, firsthand exposure to how a major global platform actually operates from stakeholder and community engagement work with Binance, one of the world's largest crypto exchanges. He is currently an AI Quality Analyst on a Google-commissioned Gemini evaluation programme delivered through Turing, performing large-scale model evaluation.
Few AI assurance practices, in Nigeria or globally, can pair that breadth of hands-on technical and platform experience with enterprise governance leadership in one person. That combination is why Afrispan's assessments look at how a system actually behaves, not only at what its policy documents claim. Steve is based in Manchester, United Kingdom, with West Africa as Afrispan's primary market and regular in-market presence planned.
Alongside its Nigerian registration, Afrispan is progressing a formal UK corporate registration, reflecting Afrispan's service reach in the UK.
Ready to see where your AI systems actually stand?
Start with a direct conversation about your system and context, not a signup form.