Personal academic website of Jecha S Jecha, PhD student at Southwest University researching artificial intelligence in education, AI adoption in higher education, and self-regulated learning. https://academic.goalpath.site/ 2026-02-01T00:00:00+08:00 Jecha S Jecha https://academic.goalpath.site/ https://academic.goalpath.site/insights/2026/02/01/beyond-hype-and-skepticism-aied-in-africa/ 2026-02-01T00:00:00+08:00 2026-02-01T00:00:00+08:00

Notes on why AI-in-education policy for Africa needs a third path between uncritical enthusiasm and blanket skepticism.

<blockquote> <p>This post reflects themes from my article <em>“Beyond skepticism: Question marks surrounding AI and AIED policies in Africa”</em> in <em>Computers and Education Open</em>. It is a short, plain-language companion to that work.</p> </blockquote> <p>Conversations about artificial intelligence in African education tend to collapse into one of two postures. The first is <strong>hype</strong>: AI will leapfrog broken systems, personalise learning for everyone, and solve teacher shortages at a stroke. The second is <strong>skepticism</strong>: AI is a distraction imported from elsewhere, unsuited to local realities and likely to widen the gaps it claims to close.</p> <p>Both postures share a flaw — they answer the question <em>in general</em>, when the honest answers are all <em>specific</em>.</p> <h2 id="the-question-marks-that-matter">The question marks that matter</h2> <p>Rather than asking “is AI good or bad for African education?”, it is more useful to ask a series of concrete questions that policy has to face:</p> <ul> <li><strong>Whose data, whose infrastructure?</strong> AIED assumes connectivity, devices, and data pipelines that many institutions do not have — and cannot assume will arrive on the timeline the technology promises.</li> <li><strong>Readiness for what?</strong> A system can be declared “AI-ready” while lacking the everyday conditions — reliable power, trained staff, maintenance budgets — that turn readiness into use.</li> <li><strong>Policy for whom?</strong> Borrowed policy frameworks encode assumptions about institutions and learners that may not hold locally.</li> </ul> <h2 id="a-third-path">A third path</h2> <p>The alternative to hype and skepticism is not a compromise between them. It is a commitment to <strong>context</strong>: policy that is explicit about local constraints, honest about what AI can and cannot do under those constraints, and evaluated against outcomes that matter to the systems adopting it. That is the direction my research on AI adoption and readiness in higher education tries to make concrete.</p> https://academic.goalpath.site/insights/2025/11/15/when-readiness-does-not-lead-to-adoption/ 2025-11-15T00:00:00+08:00 2025-11-15T00:00:00+08:00

Institutions can be ready for AI on every measurable dimension and still not adopt it. That gap is where the interesting questions live.

<blockquote> <p>This post draws on my work using UTAUT and TOE–TAM to study AI integration in higher education, including the preprint <em>“When Readiness Doesn’t Lead to Adoption.”</em> Consider it a short companion to that research.</p> </blockquote> <p>A lot of technology-acceptance research quietly assumes a straight line: measure readiness — awareness, attitudes, infrastructure, perceived usefulness — and adoption follows. In the higher-education settings I study, that line keeps breaking. Institutions score well on readiness instruments and then do not actually adopt the tools.</p> <h2 id="why-the-gap-opens">Why the gap opens</h2> <p>Readiness and adoption are measured at different levels and different moments. Readiness is often an <strong>individual, attitudinal</strong> snapshot; adoption is an <strong>organisational, behavioural</strong> outcome that unfolds over time. Between them sit factors the readiness survey never captured:</p> <ul> <li><strong>Organisational fit</strong> — does the tool match how the institution actually makes decisions, allocates budgets, and rewards staff?</li> <li><strong>Technology conditions</strong> — is there support, maintenance, and continuity, or a pilot that ends when the funding does?</li> <li><strong>Perceived risk</strong> — what happens to an educator who adopts and it fails?</li> </ul> <p>Frameworks like TOE (technology–organisation–environment) combined with TAM help name these, which is why I pair them rather than relying on acceptance models alone.</p> <h2 id="why-it-matters-for-policy">Why it matters for policy</h2> <p>If we treat readiness as a proxy for adoption, we will keep declaring success at the wrong moment — funding readiness-building while adoption stalls out of view. The more useful move is to measure adoption directly, over time, and to design for the organisational and environmental conditions that actually carry a tool from “ready” to “in use.” That is the thread my current research is following.</p>