Blind spots in Pakistan’s data governance
Written by Roshan E. Ali
Pakistan is building a framework to govern, protect and make better use of its data. But who is assessing whether this data actually represents Pakistan?
The Ministry of IT and Telecommunication released the draft National Data Governance Policy 2026, while the National Artificial Intelligence Policy was approved in July 2025.
However, the unresolved question is what methodology will ensure that the datasets defining the resulting AI systems are sufficiently representative of the people and realities they are intended to serve. Terms like “fair,” “ethical,” and “inclusive” risk becoming little more than politically correct buzzwords devoid of accountability, unless they are backed by clear standards and a defined mechanism for assessing compliance.
AI systems rely heavily on the data they use and are trained on, and we face consequences if we fail to address the practical implications of fundamental gaps and limitations in our datasets. Neither policy clearly establishes measurable standards nor a mandatory review mechanism to determine whether data sets are sufficiently representative and appropriate for informed decision-making.
If the data miss important dimensions of representation, such as the population of certain provinces, languages or forms of economic activity, among others, those omissions can be embedded in the system built on it. The quality and representation of data can therefore directly shape how accurately these systems understand, respond to, and serve the people and realities they are designed to address.
The problems in the data sets already show why these questions matter. A 2025 World Bank review of Pakistan’s National Socioeconomic Registry (NSER), one of the country’s most important databases for identifying vulnerable households, found that while 84 percent of households were registered, about 2.2 million households were excluded from the poorest 40 percent.
Exclusion from the registry previously meant exclusion from income support, disaster relief, maternal and child nutrition programs and education stipends. The evaluation also identified disparities in coverage across provinces, barriers to registration, and concerns about outdated information and data accuracy.
Without monitoring, if such data is analyzed by AI systems or used to inform AI-driven decision-making, these gaps could potentially become automated and negatively impact populations, particularly in vulnerable areas and informal sectors of the economy. Citizens whose needs and circumstances are not fully captured in digital records may be invisible to systems shaping access to health care, education and other government services.
Similarly, policies based on recorded economic activity may disproportionately reflect formally documented businesses while ignoring, for example, a woman running an unregistered home business. This risks further marginalizing citizens and economic activities that are already underrepresented in national data sets.
It also creates the risk of a negative feedback loop: populations and segments receive better-targeted services and policies, generating even more data about them, while those who are less documented receive less, remain underrepresented, and data-driven systems become even more difficult to see. With AI, incomplete national datasets will translate into socioeconomic inequality faster than we think.
Without definitive benchmarks to assess whether datasets are sufficiently representative and appropriate for their intended use, we risk making the already underserved and underrepresented even less visible. What lies ahead is the task of devising fair and measurable national standards for data collection and data quality, including clear measures of coverage, completeness, accuracy, timeliness, and representativeness that reflect Pakistan’s geographic, political, and socioeconomic realities.
Pakistan can set a strong precedent in AI and data governance by establishing an accountable mechanism to assess gaps in representation before embedding them into productive systems. Policies should also ensure that the public and private sectors cannot deploy high-impact AI systems—which could affect millions of lives—without demonstrating that the data they rely on or use meets clearly defined standards for their purpose. A country can create the perfect system, but with imperfect data, it will still produce imperfect governance.
The author is a graduate of Northwestern University with a BSc in Communication and Media Studies. He can be contacted at www.linkedin.com/in/roshaan-e-ali-067442223
The post Blind Spot in Pakistan’s Data Governance appeared first on Daily Pakistan English News.
Stay updated with the latest alerts on this story:
View Original Coverage
