AIFOD Review Warns Falling AI Costs Carry Hidden Risks for Global South

Falling artificial intelligence unit prices will not necessarily lower overall budgets and could erode essential human judgment, specialists warned during the 29th session of the Members' Weekly AI Review organised by the AI for Developing Countries Forum (AIFOD), which publishes this site. Held online on 26 September 2026, the discussion focused on the economics of AI, the Jevons paradox, and the risks of outsourcing human oversight.

AIFOD Co-Founder Tianze Zhang opened the panel by contrasting falling model expenses with overall demand, observing that increased efficiency historically expanded resource consumption rather than curbing it. Reda Sadki of The Geneva Learning Foundation emphasised that cheaper generation merely inflates verification workloads. Drawing on experiences with agentic AI deployments in West and Central Africa, he cautioned that automating foundational tasks threatens to weaken the long-term expertise needed to identify machine errors.

Vendor Lock-In and Governance

Panellists highlighted structural vulnerabilities for developing economies. Craig Gibbs from JET Education Services pointed out that budget-conscious ministries risk falling into long-term vendor lock-in with unsuitable curricula tools that they cannot afford to replace. Fabrizio Degni of Webuild added that token prices represent only a fraction of total implementation expenditures, which must cover infrastructure, governance, compliance, and cybersecurity.

Legal and economic specialists also stressed the imperative of accountability. AIFOD Senior Fellow Valerie Pelton warned that adopting cheaper systems must not lead to compromised operating standards. Professor Avtandil Gagnidze from East West University concluded that machine-scale mistakes generate substantial unmeasured costs, reiterating that human practitioners must always retain final responsibility for institutional decisions.