Glocal 2026: Building Trust in Evaluation in the Age of AI across Anglophone Africa

30 June 2026 -

During Glocal Evaluation Week 2026, held June 1–5, sessions across Anglophone Africa explored the opportunities and challenges presented by artificial intelligence (AI) and what they mean for the future of evaluation. The regional program was convened by Clear Evaluation Anglophone Africa (CLEAR-AA), an implementing partner of the Global Evaluation Initiative (GEI).

While discussions examined emerging technologies and digital innovation, a common thread ran throughout the region’s contributions: how to ensure that evaluation remains trustworthy, people-centered, and grounded in African realities in an increasingly AI enabled world. 

While discussions examined emerging technologies and digital innovation, a common thread ran throughout the region’s contributions: how to ensure that evaluation remains trustworthy, people centered, and grounded in African realities in an increasingly AI-enabled world

The regional program opened with “AI and the Future of Evaluation: Concepts, Opportunities and Emerging Debates,” where speakers reflected on how African evaluators were navigating the rapid adoption of AI tools while safeguarding principles of inclusion, equity, and contextual relevance. Discussions highlighted both the potential of AI to improve evidence generation, multilingual engagement, and learning, and the risks associated with bias, exclusion, and the marginalization of indigenous knowledge systems. Participants emphasized the importance of Africa actively shaping AI systems that reflect local contexts, languages, and development priorities.

Across the region, questions of community voice and meaningful participation emerged as another key theme. In “Algorithmic Tokenism vs. Digital Marema-Tlou: Decolonizing AI in Evaluation,” participants explored whether AI systems were genuinely amplifying community perspectives or simply transforming lived experiences into data points. Discussions highlighted the risk of algorithmic tokenism, where communities appear represented in datasets but remain excluded from interpretation and decision-making. Speakers emphasized that evaluation evidence should ultimately serve communities rather than institutions alone, and that responsible AI must be complemented by human wisdom, stakeholder validation, and continuous community engagement.

Trust and credibility in evidence also featured prominently throughout the week. Sessions such as “From Evidence to Decisions in the Age of AI: Strengthening Evaluation Use and Trust” and “Can We Trust Evaluation Findings Made by AI?” examined how evaluation systems can maintain confidence and legitimacy as AI becomes more integrated into practice. Participants agreed that while AI can support evidence synthesis, translation, and communication, it cannot replace human judgment, contextual understanding, or ethical decision-making. Discussions reinforced that trustworthy evaluation depends on transparency, appropriate levels of rigor, and meaningful human involvement throughout the evidence-to-decision process.

Related conversations focused on how evidence can be communicated and used more effectively. In “Reimagining Evidence Through Storytelling in the Age of AI,” participants reflected on the importance of ensuring that evidence is not only generated but also understood and acted upon. Discussions emphasized that credible evaluation requires strong connections between data collection, analysis, validation, communication, and implementation. Storytelling was highlighted as an important tool for translating findings into action and strengthening engagement among communities, policy makers, and other decision-makers.

Beyond practical applications, discussions also engaged with the philosophical foundations of evaluation. “The Philosophical Archetypes of Evaluation” explored questions of value, knowledge, ethics, and digital transformation, encouraging participants to reflect on how emerging technologies may influence the ways evaluators understand, assess, and communicate value in complex environments.

Together, these sessions and others reflected a coherent regional narrative. Rather than viewing AI as either a solution or a threat, Anglophone Africa's contributions highlighted the importance of responsible adoption, ethical governance, community participation, and contextual understanding. By bringing together discussions on decolonization, evidence use, storytelling, and trust, the region demonstrated that the future of evaluation will depend not only on technological innovation, but also on preserving the human relationships, local knowledge systems, and values that make evaluation meaningful.
Looking ahead, CLEAR-AA is planning a series of webinars to build on several of the week's sessions, including those the centre hosted, to continue the conversations on responsible AI adoption in evaluation.

"To ensure our practices remain grounded in African realities, the centre will spearhead the responsible adoption and ethical governance of AI tools across Anglophone Africa... we will also use storytelling to communicate data, translating evaluation evidence into transparent, trusted, and impactful actions for policy makers and local communities," said Dr Steven Masvaure, Research and Learning, CLEAR-AA.