In a joint op-ed published in The News, Federal Minister for Information Technology and Telecommunication Shaza Fatima Khawaja and UN Women Pakistan Representative Jamshed M Kazi argued that the central danger of artificial intelligence is not that machines might outthink people, but that AI systems could absorb existing societal inequalities and embed them into digital infrastructure at unprecedented speed and scale. They framed the issue as a development question for Pakistan rather than a purely technical one, one that will shape the country’s growth trajectory over the coming decade.
The authors pointed to Pakistan’s expanding digital footprint, noting the country has surpassed 200 million telecom subscribers and 150 million broadband connections, alongside record IT and IT enabled services exports of $4.6 billion last fiscal year, a 21 percent increase over the previous year. But they cautioned that this momentum does not automatically translate into inclusion, warning that AI systems trained on historical data risk reproducing patterns in which women and other marginalized groups had fewer opportunities to own businesses, build credit histories, or reach leadership positions. They cited generative AI’s tendency to depict executives, engineers, and scientists as men while defaulting to women for caregiving roles as a small but telling illustration of how these biases surface in everyday tools.
The piece cited specific figures to underscore the scale of the gap. Women make up nearly half of Pakistan’s population but hold a labour force participation rate of around 24 percent, among the lowest in the region. Globally, women hold barely 30 percent of AI jobs, with even fewer studying or working in STEM fields in Pakistan specifically. The authors noted the International Monetary Fund estimates that closing the gender gap could lift Pakistan’s GDP by as much as 30 percent, framing the exclusion of women from the digital economy as an economic cost rather than solely a social concern. They did point to recent progress, citing GSMA Mobile Gender Gap data showing Pakistan narrowed its mobile ownership gender gap from 37 to 27 percent and its smartphone gap from 48 to 30 percent in a single year, the sharpest improvement recorded among surveyed countries.
The authors outlined a four part framework for building what they termed gender responsive AI: diverse datasets, systematic bias testing, transparent governance, and design teams reflective of the societies AI is meant to serve. They said this framework underpins the newly launched Gender-Responsive AI School Pakistan, a joint initiative between the Ministry of Information Technology and Telecommunication and UN Women, supported by the Korea International Cooperation Agency, intended to bring together policymakers, academia, entrepreneurs, and technology leaders to build AI skills grounded in fairness and inclusion. They noted a UN Women review of 138 national AI strategies found only 24 mentioned gender at all, with just 18 including concrete measures to address it, positioning Pakistan’s initiative as an attempt to join a small minority of countries treating the issue seriously rather than as an afterthought.
Follow the SPIN IDG WhatsApp Channel for updates across the Smart Pakistan Insights Network covering all of Pakistan’s technology ecosystem.
