Pakistan’s Draft Artificial Intelligence (AI) Policy: Strengths, Weaknesses, and Implementation Challenges

In a world where technological innovation is reshaping industries and societies at an unprecedented rate, Pakistan stands at the threshold of a new era. The release of Pakistan’s draft National Artificial Intelligence (AI) Policy by the Ministry of Information Technology and Telecom (MoITT) in May 2023 signals a commitment by the nation to position itself competitively in the age of the fourth Industrial Revolution. The policy stems from “AI for Good” Initiative, which promote collaboration and innovation in deployment of AI solution. The goal is to ascertain that technological advancement benefits all segments of the society, which, in turn, would contribute towards realization of the Sustainable Development Goals (SDGs).

At its core, the policy delineates an ambitious strategy and inclusive framework that would allow Pakistan to participate vigorously in developing and utilising AI technologies, thereby ascertaining a share in the international technological landscape. Moreover, the policy seeks to reshape AI from a local standpoint, with an emphasis on targeting the job market, fostering skill development, and contributing towards the sustainability of the tech industry. By adopting a forward-looking approach, the policy highlights Pakistan’s resolve to not merely follow but to lead in the AI domain.

The National AI Policy outlines a holistic strategy for the integration of AI technology in Pakistan, which rests on four foundational pillars. The first pillar, “Enabling AI through Awareness and Readiness,” aims to augment public consciousness and preparedness toward AI adoption. The ultimate goal is to enhance societal understanding of AI and concurrently hone the skill set of the country’s workforce to meet the demands of this transformative technology.

The second pillar, “AI Market Enablement,” emphasises the creation of an environment conducive to the development and deployment of AI. Central to this venture is the formulation of a National AI Fund, anticipated to serve as a financial bedrock for AI-driven projects.

The third pillar, “Building a Progressive and Trusted Environment,” highlights the significance of creating a regulatory framework embedded within the ethical principles and global best practises. Through collaboration between public and private entities, Pakistan aims to generate an ecosystem that promotes research and development while ascertaining safety, security, privacy, and accountability.

The last pillar, “Transformation and Evolution,” aims to integrate AI across different sectors of society, such as healthcare, education, and governance services, to improve efficiency and productivity. The policy proposes specific action and targeted interventions like industrial transformation and sandboxing for seamless integration of AI into their operations and services.

After more than ten months of the draft release, the policy remains un-adopted. Therefore, it is crucial to identify challenges in the process of realising policy’s true potential and incorporate them in the final version. First, the policy has been standardised without the presence of dedicated data laws; the absence of a legal framework to govern data usage and protection raises doubts regarding the legitimacy and credibility of the policy.

Second, while the policy focus on achieving certain outcomes related to AI within the country, it does not taking into account the obstacles encountered in far-flung and conflict-ridden zones. This implies that the policy overrides the significance of network and communication infrastructure in these areas. For example, in remote regions of Pakistan, such as Gilgit-Baltistan, where there is inadequate conventional infrastructure, AI-driven solutions could play a crucial role in revolutionising access to healthcare, disaster management, and agricultural practises.

In addition, the draft National AI Policy does not address the paramountcy of modifying AI solutions according to the local context, thereby negating the diverse linguistics and contextual needs. For instance, in rural communities of Pakistan, where English is not the primary medium of communication, AI tutoring platforms must have the capability to deliver content in regional languages. Failure to do so could exclude a major portion of the population from assessing opportunities provided by AI.

Third, effectively implementing this policy requires credible, real-time, and quality data across various sections, including agriculture, education, health, climate, energy, etc. The availability of limited or controlled data in the current scenario could potentially disrupt the deployment of AI tools. For example, lack of access to vast medical data could hamper AI applications for disease prediction, diagnosis, and personalised treatment.

There are a few key recommendations that the Pakistani government must take into account. Essentially, the pillars of AI strategy must be rearranged in accordance with the domestic situation of the country. For example, utmost attention should be given to AI research and development, as it would eventually lay the foundation for digital infrastructure in Pakistan. Moreover, to deal with financial constraints, Pakistan must leverage platforms such as the Special Investment Facilitation Council (SIFC). To further strengthen the implementation process, a dedicated coordination unit must be formulated under the leadership of the Prime Minister. The committee must include diverse stakeholders from the National Security Division, AI experts, academicians, civil society, etc., to instruct in-line ministries to attain complete adoption of AI.

All in all, the draft National AI policy of Pakistan provides a structured approach towards integration of AI across different sector. Through collaborative and concerted efforts, Pakistan stands poised to leverage AI as a catalyst for socio-economic growth and development.