Nigeria issues warning on risks of poorly curated AI data pipelines

The Nigeria Data Protection Commission (NDPC) has raised concerns over the vulnerabilities posed by poorly curated data pipelines in developing advanced artificial intelligence (AI) models. This issue risks exposing Nigeria’s digital ecosystem to automated cyber exploitation and widespread disinformation.

The warning was delivered during a Stakeholders’ Consultation on Frontier AI hosted by the Nigeria IT Hub (NITHUB), University of Lagos, and the Tech Governance Project. Representing the NDPC, Abubakar Mahmud from the IT and Cybersecurity Department emphasized the potential for malicious actors to exploit gaps in AI data governance.

At the centre of the discussion was the role of foundation models—AI systems trained on vast datasets to perform diverse tasks such as text and image generation, software development, and data analysis. The commission highlighted that the quality, origin, and management of training data fundamentally impact the reliability and security of these AI models.

The NDPC referenced the Nigeria Data Protection Act, 2023, as a vital legal framework to mitigate these risks. Key principles such as data minimisation—which limits data collection to what is necessary—and algorithmic transparency, which demands clear understanding of AI decision-making processes, were underscored as essential safeguards.

Furthermore, the Commission advocated the inclusion of the international standard ISO/IEC 42001:2023 into the country’s forthcoming AI Regulatory Sandboxing Platform. This standard outlines best practices for AI governance, aiming to institute privacy protections and accountability mechanisms from the earliest stages of AI development.

The proposed regulatory sandbox will provide a controlled environment where AI technologies can be tested under government oversight before wider deployment. Such frameworks aim to balance innovation stimulation with risk management.

The NDPC also called for broad collaboration among government agencies, academia, industry stakeholders, and civil society. This multi-stakeholder approach is intended to facilitate co-creation of policies, promote regional alignment across African nations, and maintain consistency with international standards.

This development is significant for Nigeria as AI technologies increasingly influence the economy and public service delivery. Ensuring data protection in AI is crucial to preventing digital exploitation and maintaining public trust in emerging technologies.