Digital Inequality
Digital Inequality: Artificial intelligence is no longer a distant idea. It already shapes our daily lives. We use it in our phones, banks, and schools. Many systems now use AI to make decisions about people. Experts often call AI the “new electricity.” However, electricity can reach everyone equally. AI does not always do that.
Automated Inequality
Governments around the world now use AI in governance. Pakistan is also moving in this direction. At Indus AI Week 2026, policymakers, business leaders, and technology experts discussed AI’s role in national growth. Pakistan’s National Artificial Intelligence Policy 2025 also outlines plans to train people, strengthen research, and promote responsible AI use.
Unequal Access
Yet technology does not benefit everyone equally. Consider Shazia, a woman who lives in a small village. She checks messages only when he is home. She does not appear in digital records as an independent user. Still, new automated systems in taxes, social benefits, jobs, and public services aim to serve people like her.
Gender Digital Gap
We must remember women like Shazia when we discuss AI and digital progress. AI systems learn from existing social data. If society contains inequality, AI can repeat it. Algorithms can copy these biases without intention.
Statistics show a clear gap. The UNDP National Human Development Report 2023 shows that only 26 percent of women in Pakistan use the internet. In comparison, 47 percent of men use it. The gap is wider in rural areas. Only 7 percent of rural women go online. Many women say male family members control their phone access.
Data Gaps
These numbers matter because AI depends on data. It learns from online activity, financial transactions, and service use. When digital data does not capture women’s lives, systems fail to understand their realities. Over time, these gaps become part of the system.
Pakistan’s AI policy promotes inclusion. It promises to expand AI training for women and underserved groups. These steps are important. However, training alone is not enough. Women must also help shape AI policies and decisions.
Responsible AI
Responsible AI needs strong rules and safeguards. Governments must check systems for bias. They must study their impact on gender equality. Oversight bodies should include women and gender experts. As governments automate decisions, power shifts into computer code. If designers ignore gender perspectives, systems may create long-term exclusion.
Economic Inequality
This problem is not only technical. It reflects governance choices. It also affects the economy. The World Economic Forum reports that people with generative AI skills can earn up to 36 percent higher wages. Yet UNESCO reports that women hold only about 22 percent of AI-related jobs worldwide. If Pakistan grows its digital economy without closing this gap, inequality may increase.
Algorithmic Bias
Technology already shows signs of bias. MIT researchers studied facial recognition systems. They found that some systems worked less accurately for darker-skinned women than for lighter-skinned men. This happened because the training data lacked diversity. Technology reflects the data it receives.
Pakistan already faces challenges in AI governance and data coordination. If policymakers ignore gender in early planning, they may struggle to fix problems later. Once institutions adopt automated systems, corrections become costly and difficult.
Time to Act
However, Pakistan still has time to act. Policymakers can include gender considerations before systems expand widely. They can review datasets carefully.
Shazia may never attend an AI conference. She may never learn to code. Yet the systems we design today will shape her life. They will influence the services she receives, the benefits she qualifies for, and the risks she faces.
AI Choices
Pakistan will continue to adopt AI. The real question is how it will use it. The country can allow technology to increase inequality. Or it can build strong systems that prevent it. Once inequality enters automated systems, removing it becomes much harder.
Published in Dawn, 8th March 2026
