Pakistan is moving closer to formal Artificial Intelligence legislation in Pakistan as lawmakers, regulators, and industry experts push for a clear legal structure around AI use. The Regulation of Artificial Intelligence Act, first introduced in the Senate in 2024, remains under review, while the National AI Policy 2025 has already been approved by the federal cabinet. Together, these developments mark the country’s first serious attempt at federal legislation on AI, aligning with global frameworks such as the AI Act Regulation adopted by the European Union.
Background
For years, Pakistan lacked any dedicated legal framework to govern artificial intelligence. Data protection, algorithmic accountability, and AI-driven decision-making existed in a legal grey area. As AI tools became common across banking, healthcare, education, and public administration, the absence of proposed AI legislation created growing concern among legal experts and technology policymakers.
The push for reform gained momentum after global regulators, particularly the European Union, finalized the AI Act Regulation under EUR-Lex, setting a precedent that many countries, including Pakistan, are now trying to follow. This shift encouraged local lawmakers to introduce a dedicated Artificial Intelligence Act aimed at bringing order to an otherwise unregulated digital space.
Details of the Proposed Legislation
The centerpiece of Artificial Intelligence legislation in Pakistan is the “Regulation of Artificial Intelligence Act 2024,” introduced in the Senate by lawmaker Afnan Ullah Khan. This bill proposes the creation of a National Artificial Intelligence Commission based in Islamabad, tasked with overseeing AI deployment across critical sectors such as health, education, defense, and finance.
Under the draft law, organizations found violating AI safety or transparency requirements could face fines ranging from Rs 1.5 billion to Rs 2.5 billion, depending on the severity of the breach. The bill also proposes mandatory human oversight for AI systems used in sensitive decision-making, a provision similar to risk-based classifications found in the AI Act latest version PDF published by EU regulators.
Alongside this pending bill, the government approved the National AI Policy 2025, a broader strategic document built on six pillars covering innovation, awareness, security, sectoral transformation, infrastructure, and international partnerships. While the policy itself is not binding law, it sets the direction that future federal legislation on AI is expected to follow.
Anyone researching this topic can locate the original Senate submission through the official Artificial Intelligence Act PDF hosted on the Senate of Pakistan’s website, which outlines the bill’s full text, structure, and proposed penalties in detail.
Quotes and Expert Opinion
Legal analysts tracking the bill through various AI legislation tracker platforms note that Pakistan’s proposed framework borrows heavily from international models, particularly the structure used in the AI Act Regulation. Policy researchers have pointed out that Pakistan’s existing laws, including sections of the Penal Code and Evidence Act, were never designed to address issues like automated decision-making or AI-generated evidence, making updated legislation necessary rather than optional.
Officials associated with the Ministry of Information Technology and Telecommunication have described the National AI Policy 2025 as a foundational step meant to prepare the country for a technology-driven economy, while acknowledging that a dedicated regulatory law is still required to give the policy legal force.
Regional and Global Impact
Pakistan’s approach mirrors a broader global trend. The European Union’s AI Act, formally known as Regulation (EU) 2024/1689, has already introduced risk-based categories for AI systems and set compliance deadlines extending into 2027 and 2028 following recent amendments. Countries across Asia, including India, Bangladesh, and Sri Lanka, are simultaneously drafting their own frameworks, creating a competitive regional environment around responsible AI governance.
For Pakistan, passing a functional Artificial Intelligence Act could improve investor confidence, attract foreign technology partnerships, and strengthen the country’s position in future trade discussions involving digital services. On the other hand, delays in finalizing federal legislation on AI could leave gaps in data protection and consumer safety, especially as AI tools become more common in everyday financial and healthcare services.
Challenges Ahead
Despite the progress, several obstacles remain before Pakistan can finalize comprehensive AI legislation. Infrastructure limitations, inconsistent internet access, and a shortage of trained regulators are frequently cited as major hurdles. Experts following the AI legislation tracker closely note that the Senate’s Standing Committee on Information Technology continues to review the 2024 bill, with revisions expected to narrow penalty ranges and clarify definitions around “critical sectors.”
There are also concerns about overlapping jurisdiction between the proposed AI Commission and existing bodies such as the Pakistan Telecommunication Authority. Resolving these overlaps will likely determine how quickly the proposed AI legislation moves from committee review to a full parliamentary vote.
Conclusion and Future Outlook
Artificial Intelligence legislation in Pakistan is still a work in progress, but the direction is increasingly clear. With the National AI Policy 2025 already approved and the Regulation of Artificial Intelligence Act under active Senate review, the country appears committed to building a formal regulatory structure around AI use. As global frameworks like the EU’s AI Act Regulation continue to evolve, Pakistan’s lawmakers are expected to keep refining their own bill to strike a balance between innovation and public safety.
Industry stakeholders and legal observers will be watching closely to see whether the Senate committee finalizes the bill in its current form or introduces further amendments before it reaches a full vote.
Frequently Asked Questions
What are the 7 principles of the AI Act?
While different frameworks phrase them slightly differently, the core principles generally referenced under AI Act-style regulation include human oversight and accountability, transparency in how AI systems reach decisions, fairness and non-discrimination in outcomes, data privacy and security protections, robustness and technical safety of AI systems, traceability of AI-generated content and decisions, and a risk-based classification system that applies stricter rules to high-risk AI applications such as those used in healthcare, law enforcement, or financial services. These principles form the backbone of most modern AI regulation efforts, including discussions around Pakistan’s own proposed legislation.
What are the 4 types of AI?
AI systems are commonly grouped into four broad categories. Reactive machines respond to specific inputs without using past experience, such as basic recommendation engines. Limited memory AI can reference recent data to improve decisions, which describes most AI tools in use today, including chatbots and predictive models. Theory of mind AI, still largely theoretical, would understand emotions and intentions to interact more naturally with humans. Self-aware AI, the most advanced and currently hypothetical category, would possess its own consciousness and understanding of its own state. Most real-world regulation, including proposed AI legislation in Pakistan, currently focuses on reactive and limited memory systems since these are what businesses and governments actually deploy.
What is the AI legislation 2026?
As of 2026, AI legislation around the world is entering a more mature phase. In the European Union, the AI Act Regulation under EUR-Lex has seen targeted amendments adjusting compliance timelines, with high-risk system obligations now phased in through 2027 and 2028. In the United States, a national policy framework proposal was released in early 2026 outlining seven legislative goals for federal AI oversight. In Pakistan, 2026 has brought continued Senate committee review of the Regulation of Artificial Intelligence Act 2024, alongside implementation steps tied to the National AI Policy 2025, though a fully enacted federal AI law has not yet been passed.
