Rabia Mustafa
In a recent case titled Ishfaq Ahmed (Petitioner) v. Mushtaq Ahmed, etc. (Respondent(s)), Syed Mansoor Ali Shah, the Honourable Judge of the Supreme Court of Pakistan explained in detail of using Artificial Intelligence as an innovative road in the field of judiciary, its limitations and future considerations.
The said judgment emphasizes the urgent need to address systemic delays in Pakistan’s judiciary, especially at the district level where most disputes arise and backlog pressure is greatest. While structural reforms must continue, the current crisis demands immediate, innovative solutions. One such solution is the integration of Artificial Intelligence (AI) into court and case management systems. However, this must be done within constitutional limits, specifically under Articles 10A and 37(d) of the Constitution, which guarantee the right to fair and speedy justice. If carefully implemented, AI can enhance access to timely justice and reduce systemic backlogs.
The Superior Court explains the growing global interest among judges in using commercial Generative AI platforms like ChatGPT, Copilot, and DeepSeek for legal research, drafting, and decision-making. Notable examples include a Colombian judge citing ChatGPT in a ruling (2023), appellate judges in Washington D.C. referencing it in official opinions (2025), and a Pakistani judge using it in both civil and criminal cases, sparking debate over the appropriateness of AI in judicial decisions. Given its potential to improve judicial efficiency, Syed Mansoor Ali Shah argues that this is an opportune moment to explore AI’s role in the judiciary to help fulfill the constitutional guarantees of fair and expeditious justice.
The Use of AI to Enhance Judicial and Institutional Efficiency
The Honorable Justice discusses how Artificial Intelligence (AI), when applied within principled boundaries, can significantly boost judicial productivity and efficiency without replacing human judgment. AI can assist judges by enhancing functions like legal research, drafting, and decision support.
Key Characteristics of AI
Smart Legal Research: AI can swiftly analyze extensive legal databases to provide judges with relevant precedents, laws, and scholarly commentary. Various AI Tools are increasingly used to improve research and drafting efficiency, accuracy, and clarity. For example, the Federal Judicial Academy (FJA), Islamabad, has introduced ‘Judge-GPT,’ an AI tool developed with ETH Zurich. Tailored specifically for Pakistan’s legal system, it is currently being used by about 1,500 judges in the District Judiciary. Initial feedback suggests it helps speed up legal research and improves the quality of court orders, allowing judges to focus more on substantive legal issues.
Language Precision and Drafting: AI-powered tools help improve grammar, syntax, and structure, resulting in clearer, more coherent, and professional judicial writing.
Comparative Jurisprudence: AI enables efficient access to international legal databases, allowing judges to engage with foreign jurisprudence and promote the evolution of domestic legal thought through interaction with comparative and international law.
Decision-Making Support: AI tools assist in managing complex case files, pinpointing critical legal questions, and summarizing large volumes of records, supporting judges in making better-informed and timely decisions.
Consistency and Coherence: AI can enhance consistency and coherence in judicial decisions by identifying conflicting precedents and promoting uniformity in jurisprudence. This contributes to the predictability and stability of legal outcomes, strengthening public trust in the judiciary. However, it is crucial to recognize that AI should remain a supplementary tool, not a replacement for judicial reasoning, discretion, or interpretive judgment. The judges must maintain full responsibility for ensuring the accuracy, ethical integrity, and confidentiality of their decisions, with AI serving as an aid to, not a substitute for, human judgment.
Administrative Efficiency: The importance of case allocation in ensuring administrative efficiency within the judiciary cannot be denied. Traditionally managed by court administrators or senior judges, case allocation can sometimes raise concerns about partiality, judge shopping, or unequal caseload distribution. To address these issues, many judicial systems have adopted AI-driven mechanisms to promote objectivity and transparency. In essence, AI-driven case allocation systems aim to reinforce procedural fairness by eliminating discretion, preventing bias, and promoting efficiency in judicial processes.
Gaps in the AI Systems – A Warning
The judgement also highlights several ethical concerns and gaps associated with the use of AI in the judiciary, urging the need for a careful, principled approach to its adoption:
Transparency and Explainability: AI models, particularly deep learning systems, often operate as ‘black boxes,’ meaning their internal decision-making processes are unclear or too complex to interpret. This lack of transparency can undermine public trust, as AI-generated outcomes may not be easily justified, leading to accountability issues.
Hallucination: AI systems like ChatGPT can produce fabricated or incorrect outputs (known as ‘hallucinations’) because they rely on statistical pattern-matching rather than fact-checking. This can result in the AI confidently citing non-existent cases or procedural facts, eroding trust in legal processes. Courts have started requiring attorneys to verify AI-generated text before filing to prevent such issues.
Accountability and Human Oversight: AI must always remain a supplementary tool in judicial proceedings, with human oversight ensuring that judges retain full accountability for decisions. AI may assist by suggesting outcomes or drafting documents, but judges must retain ultimate responsibility and authority. A ‘human-in-the-loop’ approach ensures that any questionable AI outputs can be corrected and that judicial integrity is preserved.
Fairness and Non-discrimination: AI systems can perpetuate or amplify biases, particularly when trained on historical data that reflects societal inequalities. This is especially concerning in areas like bail and sentencing decisions, where AI might unintentionally replicate or worsen existing biases. AI adoption in the judiciary must be designed to avoid reinforcing discrimination, ensuring impartiality and fairness in legal determinations.
Artificial Intelligence, Judging & Fair Trial
The right to a fair trial and the implications of AI in judicial decision-making are extremely important. AI must not compromise the core principle of judicial autonomy. Although AI has the potential to improve the consistency and efficiency of legal processes, there is concern about the risk of biases and the potential for limiting judicial discretion. AI-assisted rulings must adhere to the principles of fairness and transparency. Several international bodies, including the European Commission for the Efficiency of Justice, Council of Europe, and the EU, advocate for the ethical deployment of AI in judicial contexts. While AI can assist in tasks such as legal research and case management, it cannot replace the critical human role in applying nuance, moral judgment, and equity.
A Caution while Judging with AI
Moreover, the judgement strongly asserts that AI must never replace judicial decision-making. It emphasizes that adjudication is a constitutional function deeply rooted in legal reasoning, institutional independence, and human empathy—qualities that AI cannot replicate. The essence of justice lies in the moral and emotional aspects of decision-making, such as compassion, ethical discernment, and the understanding of human suffering, which AI currently cannot comprehend. AI should be viewed as a supportive resource, not as a replacement for human reasoning and deliberation. The dignity of the judicial role lies in its deliberative process, which remains inherently human.
Judiciaries in countries such as the USA, UK, Canada, Australia, and Singapore are gradually integrating AI-driven tools into legal research and judicial workflows, though with careful oversight and limitations.
The Future of AI
The judgment shed light on the future of AI integration in the judicial system, highlighting both its potential for reform and its limitations. It emphasizes that as AI technologies rapidly evolve, their impact on the judicial system remains uncertain, and they may eventually address some of the gaps currently identified. However, certain core values must always be preserved to ensure the justice system remains fair and just.
Human Dignity and Compassion must remain central to the judicial process, ensuring that AI never overrides mercy or individualized consideration in judgments. AI systems must be designed to prevent biases and ensure equal treatment under the law. The technology should be programmed to avoid replicating historical injustices and should be built with anti-discrimination principles. Human judges must retain ultimate authority in interpreting legal and moral standards, ensuring that AI supports but does not replace judicial reasoning. The justice system should still allow for punishment that serves societal healing, not just efficiency. AI must not overshadow restorative justice or the rehabilitation of offenders. As AI continues to evolve, its role must be framed by ethical and legal guardrails to ensure that it enhances, rather than diminishes, justice.
The Presence of AI in Mediation and ADR
The evolving role of AI in the legal profession, particularly its ability to automate tasks like legal research, case summaries, contract review, drafting, and argument building is explored. While AI excels in logical and fact-based aspects of legal work, it raises an important question: What will be left for lawyers to do?
Despite AI’s increasing capabilities, several key functions remain irreplaceable by AI, particularly in the area of mediation and alternative dispute resolution (ADR). These processes rely heavily on human qualities such as emotional intelligence, empathy, and human interaction As AI continues to handle more repetitive and fact-based tasks, lawyers can pivot toward mediation and ADR, positioning themselves as experts in these fields.
Ultimately, while AI transforms many aspects of legal practice, the future of dispute resolution may lie in mediation, where lawyers can offer value that AI cannot replace.
Conclusion
In the end, the judgment concluded,
“AI must be welcomed with careful optimism. It can streamline judicial functions, reduce delays, and expand access to legal knowledge. But it cannot replicate the moral, ethical, and empathic reasoning that lies at the heart of judging.53 Courts must thus pursue a calibrated integration harnessing AI’s efficiencies without surrendering the conscience, independence, and humanity that justice demands.”
The Superior Court recommended,
“The National Judicial (Policy Making) Committee in collaboration with the Law and Justice Commission of Pakistan consider developing comprehensive guidelines on the permissible uses of AI within the judiciary. These must delineate clear boundaries, ensuring that AI is used only as a facilitative tool and never in a manner that compromises human judicial autonomy, constitutional fidelity, or public trust in the justice system. Let a copy of this judgment be dispatched to both the Law and Justice Commission of Pakistan and the National Judicial (Policy Making) Committee for preparing guidelines to regulate this emerging intersection of law and AI.”
The writer is Senior Research Fellow at the School for Law and Development
