This executive summary lays out highlights from the report Inclusion in Practice: Governing AI-Enabled Service Delivery in Bangladesh written by Max Bell School Master of Public Policy students Andre Policzer, Jehan Mohamed, and Kimberley Moriarity as part of the 2026 Policy Lab.
Access the summary and presentation below, and read their full report here.
Nations across South Asia are investing rapidly in artificial intelligence (AI) and digital public infrastructure (DPI) to deliver government services and drive economic growth. The expansion of AI and DPI capabilities could be transformative for government service delivery to expand access to government benefits, welfare supports, and digital identification. But this transformation is unfolding in a setting marked by structural inequality and limited regulatory capacity with emerging technologies. Navigating this presents a governance challenge that cannot be reduced to solving technical problems. The language of ethical and inclusive AI can be easily invoked for government credibility, but an AI system can only ever be as good as the data and design choices beneath it.
This report responds to the challenge posed by BacharLorai Global: how can South Asian nations govern the rise of AI and digital public infrastructure in ways that protect rights, strengthen democratic institutions, and accelerate inclusive economic growth? We’ve refined this regional scope into an actionable report centered on Bangladesh that asks how to govern AI-enabled digital government service delivery to promote inclusivity alongside economic growth? Our insights into this question directly frame our research and policy insights in this report.
Taken together, Vision 2041 and the recent draft National AI Policy (2026-2030) mark Bangladesh’s clearest policy acknowledgement to date that rapid digitalization must be matched by a more deliberate framework for AI governance.1 In the wake of the transition to the new government, the regulatory environment is still nascent, and there is a clear policy window to design frameworks centred on accountability, transparency, and equitable access, before market-driven incentives of efficiency-first design choices lock in inequitable outcomes.
This report is intended to strengthen and build upon the National AI Policy as it moves toward implementation through three actionable policy recommendations:
- Algorithmic Impact Assessments (AIAs): Implement audits of government service delivery tools through the use of pre- and post-deployment AIAs for automated decision-making to ensure administrative actions adhere to the highest standards of accountability, transparency, and mechanisms for citizen redress.
- Procurement Standard for Bangla and Minority-Language Operability in AI-enabled services: Enable government procurement to mandate AI systems’ operational fluency in Bangla, regional dialects, and minority languages to better target low-literacy users, with reporting requirements disaggregated by accuracy and usability to mitigate risks of exclusion.
- Informal Access Brokers: Leverage existing social and civil society networks to support digital adoption by integrating trusted community intermediaries into AI service delivery for awareness, literacy, access, and community feedback, to reach populations least likely to engage with digital systems directly.
Download the full version of this report here.
Authors: Andre Policzer, Kimberley Moriarity, and Jehan Mohamed
