PIDE calls for trust, accountability in Pakistan’s AI, fintech future

ISLAMABAD, Aug 4 (APP):The Pakistan Institute of Development Economics (PIDE) on Tuesday called for building Pakistan’s artificial intelligence (AI) and fintech ecosystem on trust, inclusion and institutional accountability, stressing that responsible technology adoption should prioritise measurable public value, human oversight and evidence-based policy-making.

The institute said Pakistan’s digital transformation should focus on responsible AI adoption and inclusive fintech development rather than technology for its own sake, with stronger governance, transparent data practices and clear human accountability underpinning future policy and institutional reforms.

These views were expressed during a seminar titled “Landscape of AI and Fintech in Pakistan,” organised by PIDE to examine how artificial intelligence, financial technology and big-data analytics could strengthen economic policymaking while safeguarding public trust, institutional integrity and sensitive information.

Speaking on the occasion, Associate Professor of Artificial Intelligence at the University of East London, Dr Mustansar Ali Ghazanfar, said AI and fintech should be guided by three interconnected principles, including value, intelligence and accountability to ensure sustainable institutional benefits.

He said AI could accelerate policy research and evidence gathering, while financial and administrative data could provide deeper insights into economic activity.

However, he emphasised that these gains would be meaningful only if supported by verified evidence, clearly defined ownership and human oversight.

Dr Ghazanfar said institutions should identify their operational needs before adopting AI technologies by determining where they could create measurable value, the data required, the safeguards needed and who would remain accountable for AI-generated outputs.

He observed that predictive, generative and agentic AI systems involved varying levels of risk and autonomy, requiring corresponding levels of oversight based on data sensitivity and the potential consequences of errors.

Calling for a shift from fragmented AI use to governed institutional workflows, he said organisations should adopt approved tools, maintain documented evidence, ensure human review and regularly evaluate performance.

Describing AI transformation as primarily an institutional and cultural process rather than a purely technological exercise, he stressed that people, processes and data remained central to successful digital reforms.

Highlighting Pakistan’s fintech sector, Dr Ghazanfar acknowledged improvements in digital-payment infrastructure but said progress should be measured not merely by transaction volumes or account registrations, but by active usage, affordability, consumer confidence, effective complaint mechanisms and improved financial resilience.

He noted that challenges in cross-border payments, digital readiness and financial inclusion persisted, particularly for exporters, freelancers, women, rural communities and low-income groups.

Warning against the misuse of public AI platforms, he advised institutions against uploading confidential research, financial records, internal documents, or personal information, and recommended data classification and secure environments for sensitive material. While AI could assist researchers in finding sources, organising evidence and preparing initial drafts, he said human experts must verify factual claims and retain final decision-making authority.

To encourage responsible AI adoption, Dr Ghazanfar proposed that PIDE launch a structured 90-day pilot project to test AI-assisted research workflows before considering wider implementation.

The seminar focused on building trusted digital services, evidence-based policy-making and accountable AI governance rather than pursuing technology solely for visibility or scale.

It also called for promoting private-sector innovation through effective regulation, stronger public-private partnerships and practical AI applications suited to Pakistan’s available resources.