Digital literacy is frequently offered as an immediate response to debates on the challenges posed by artificial intelligence (AI), digital platforms, and emerging technologies. Yet, prevailing policy responses remain dominated by expansive and contested regulatory frameworks that tend to prioritise restrictions on freedom of speech and expression over ethics-based engagement with the evolving digital landscape.
Policy experts argue that frameworks which fail to account for AI’s impact on entrenched gender-based inequalities and systemic discrimination, particularly those limiting access to rights, safety, and opportunities for women and vulnerable groups, are unlikely to deliver meaningful or lasting social change.
These concerns are underscored in a recent study by the Digital Rights Foundation titled Emerging Technologies in Pakistan: Towards a People-Centred Policy Framework. The research examines Pakistan’s rapid adoption of technologies such as artificial intelligence, cryptocurrency, and advanced connectivity systems, including 5G and satellite internet.
While recognising innovation-driven growth, the report highlights serious regulatory gaps, privacy risks, censorship pressures, algorithmic bias, and the escalation of technology-facilitated gender-based violence (TFGBV). It calls for governance approaches grounded in human rights rather than innovation-first policymaking alone.
The report urges alignment with international human rights standards while stressing the importance of grounding policy in local social and legal realities. It flags AI-driven misinformation, weak legal remedies for digital harms, and the disproportionate burden these harms place on women. It urges government and policymakers to adhere to UNESCO’s Recommendation on the Ethics of Artificial Intelligence (2021), adopted by all 193 Member States, which emphasises human rights, transparency, fairness, and human oversight as principles that must guide the entire AI lifecycle.
Two key issues emerge from the research. First is the urgent need for people-centred AI governance frameworks capable of addressing algorithmic bias, AI-generated misinformation, surveillance, and privacy threats, particularly for women, adolescent girls, children, transgender persons, and other vulnerable groups, while safeguarding constitutionally guaranteed fundamental rights.
TFGBV often persists with greater impunity than offline gender-based violence
Second is the outsized impact of AI on women in contexts marked by weak legal frameworks and enforcement gaps, most visibly through TFGBV. Citing Kate Crawford’s Atlas of AI: Power, Politics, and the Planetary Costs of Artificial Intelligence, the report reinforces the idea that AI systems are deeply embedded in political and economic structures and are often designed to serve dominant interests, rendering artificial intelligence a modern registry of power.
In Pakistan, public discourse on AI frequently swings between optimism and alarmism, shaping debates on freedom of expression and TFGBV. In a society shaped by entrenched socio-cultural and socio-religious norms, this concentration of technological power risks reinforcing mechanisms used to regulate and control the lives and choices of women, girls, and transgender persons. Gender-based inequalities that historically fuelled physical violence are increasingly replicated online through TFGBV, facilitated by anonymity and the absence or weak enforcement of effective legal protections.
Women in public-facing roles, including politics, journalism, media, entertainment, and social media, are among the most affected. Digital harassment, coordinated trolling, violent threats, morphed images and videos, disinformation, and fake news have become routine. These abuses have been linked to social ostracisation, severe mental health consequences, and, in extreme cases, loss of life. Notably, researchers point out that TFGBV often persists with greater impunity than offline gender-based violence.
The report also observes that Pakistan’s technology policy landscape remains focused on catching up rather than shaping the direction of responsible innovation. The National Artificial Intelligence Policy 2025 prioritises innovation and capacity building but omits critical safeguards, including comprehensive data protection laws, enforceable privacy standards, transparency in algorithmic decision-making, and gender-sensitive provisions.
The absence of any reference to the draft Regulation of Artificial Intelligence Bill, 2024, further weakens the policy’s credibility. Despite mounting evidence that women and marginalised communities face disproportionate risks from AI-driven surveillance and content regulation, the policy lacks an intersectional lens.
In the absence of personal data protection legislation, and alongside contested laws such as the Prevention of Electronic Crimes Act, 2016 (PECA), Pakistan’s national AI framework remains ill-equipped to protect human rights or address gender-based vulnerabilities.
Pakistan’s approach to emerging technologies can best be described as adaptive rather than transformative. The country is unlikely to benefit meaningfully from the Fourth Industrial Revolution if policy processes continue to replicate global models without local contextualisation, prioritise institution-building over accountability, or rely on criminal law and speech restrictions as primary regulatory tools. Such techno-solutionist approaches implicitly accept global power imbalances while portraying AI as an unqualified good, sidelining concerns around rights, accountability, and gendered harms, particularly TFGBV.
The report ultimately calls for a multilayered, gender-responsive strategy aimed at reducing harm to women and vulnerable communities while addressing AI-enabled abuse, from hate speech to TFGBV. It advocates for policies built through participatory, ground-up processes that shift decision-making power toward those most affected—women, gender-diverse individuals, religious and ethnic minorities, journalists, and human rights defenders.
The guiding values proposed include solidarity through collective action, caution in deploying high-risk technologies until safeguards exist, intersectionality in centring marginalised experiences, and meaningful participation through co-design with binding influence rather than symbolic consultation.