Pakistan governs artificial intelligence through general law and executive policy, not through a statute written for it. The live AI disputes here are criminal, deepfakes prosecuted under the cyber-crime code, rather than questions of ownership.
A note that is itself the finding: no instrument written specifically for artificial intelligence binds anyone in Pakistan today. The binding tier is general law that AI conduct happens to meet; the AI-specific texts are all either policy or unfinished.
The foundational federal cyber statute: criminalizes unauthorized access, electronic defamation, hate speech and online fraud; empowers the telecom regulator to block unlawful content and the investigation agency to act. Applies to anyone using an information system in Pakistan. Not AI-specific; it addresses neither synthetic media nor AI operators.
Source: pakistancode.gov.pkInserts Section 26A, criminalizing intentional dissemination of false or fake information likely to cause fear, panic or disorder, up to three years and a two million rupee fine, with the terms left undefined. Creates a Social Media Protection and Regulatory Authority that registers platforms and orders removal of unlawful or offensive content, plus complaint councils and tribunals, and replaces the cyber-crime wing with a National Cyber Crime Investigation Agency. It is a false-information and content-moderation statute; it contains no deepfake, synthetic-media or AI-labeling provision.
Source: na.gov.pk gazette; analysis rsilpak.orgAn enabling statute for national digital transformation. Establishes the Pakistan Digital Authority, the National Digital Commission and a strategic oversight committee to build a national digital masterplan, interoperable public infrastructure and a digital identity layer, with the authority able to issue binding standards. It is a digital-infrastructure governance statute, not a data-protection law, and creates no AI-specific operator duty.
Source: na.gov.pk gazette; pakistancode.gov.pkThe central AI instrument and a policy, not a statute. A six-pillar roadmap for an ethical and inclusive AI ecosystem: an AI Council and a regulatory directorate under the ministry, a national AI fund, centres of excellence, regulatory sandboxes and sectoral roadmaps, with a target of a million trained professionals by 2030. It recommends future deepfake-labeling and disinformation rules but imposes no enforceable duty, and no implementing regulations with sanctions have followed.
Source: moitt.gov.pkA governance framework for protecting national information infrastructure across public and private institutions, with incident-response structures and security baselines. Strategic and voluntary; no direct sanctions. Included as the security backdrop AI systems operate against.
Source: moitt.gov.pkRequires public-sector entities to prioritize cloud solutions for new technology investments, with a cloud office, a provider-accreditation regime and data-classification baselines. Binds government procurement practice as policy rather than as a duty-creating statute.
Source: moitt.gov.pkThe umbrella digital-economy strategy covering connectivity, skills, innovation and e-government. Relevant as the soft-law backdrop to the later AI and digital instruments; no AI-specific content of its own.
Source: moitt.gov.pkWould establish a comprehensive data-protection regime, with consent, data-subject rights, cross-border transfer limits and a national data-protection commission. Long pending across drafts from 2020 onward, debated again in a Senate committee in early 2025, and still not passed by both houses. Its absence leaves Pakistan without a general data-protection statute for AI processing to meet.
Source: moitt.gov.pkWould set a registration and risk-control framework for algorithmic trading in capital markets to manage flash-crash and manipulation risk. The nearest thing in the pipeline to a binding duty on automated systems, though algorithm-specific rather than AI-general.
Source: SECP consultationFlagged in the central bank's 2024 financial-stability review as being finalized: guidance to foster trust, transparency and accountability in AI-driven financial services. Not yet a binding circular.
Source: SBP Financial Stability Review 2024Founding member; signed Shanghai, 16 Jul 2026, by the Deputy Prime Minister and Foreign Minister. Official source: China State Council and Xinhua releases naming Pakistan among the 29; acknowledged in Pakistani reporting.
26 May 2026; records Pakistan supporting the establishment of WAICO. Official source: China embassy Islamabad joint-statement text.
Adopted by consensus 22 Sep 2024; applies to Pakistan as a UN member. Official source: A/RES/79/1 annex, un.org.
Adopted 23 Nov 2021, as a UNESCO member; national readiness assessment not yet completed. Official source: unesco.org; UNESCO Pakistan governance-forum record.
Consensus adoption 1 Jul 2024; Pakistan a named co-sponsor. Official source: UN records; China UN mission co-sponsor list naming Pakistan.
Consensus adoption 1 Jul 2024; Pakistan a named co-sponsor. Official source: UN records; China UN mission co-sponsor list naming Pakistan.
Full member; Astana Declaration of 9 Jul 2024 carries an AI-cooperation paragraph. Official source: sectsco.org summit page.
Full member; the 2025 Tianjin summit adopted AI-cooperation commitments and a roadmap. Official source: sectsco.org summit page (exact document title to be quoted only from the SCO original).
Axes: specificity (x, 0-100) is how squarely the instrument is written for AI; force (y, 0-100) is how hard it binds. Placements are editorial judgments for the comparison plot; the tier column is the authoritative classification. The visible pattern is a gap: the instruments that bind sit low on specificity, and the instruments that are specific sit low on force.
Tap a marker
The binding and prescriptive corner is the crowded one: the CAC-led measures put force and detail together, service by service. The sector guidance sits just left of the binding line, nonbinding in form but treated as obligatory. Dashed markers are drafts; the National AI Law would consolidate the whole stack into one statute.
Four matters qualify under the inclusion test, with AI materially at issue as subject, tool or evidence. Pakistan publishes few neutral citation numbers for district-court and investigation-agency matters; none of the matters below carries one, and per the fabrication screen none is invented. The centre of gravity is criminal: synthetic media prosecuted under the cyber-crime code, not intellectual-property disputes, of which none was found.
Portals and sweeps: WIPO Lex and the IP office pages; the Pakistan Code; superior-court reported-judgment pages; and Pakistani media of record (Dawn, The News, Express Tribune, ProPakistani, Courting the Law). The subscription case database was not accessed.
Pakistan's copyright and patent statutes predate the question: a 1962 ordinance with no computer-generated-works provision, a closed fair-dealing list, and no office guidance or disclosure duty behind them.
The AI-IP Index reads a country on two questions that usually get folded together. Accommodation asks how far the law will go to protect intellectual property that a machine helped produce. Institution asks how much administrative machinery sits behind that law: its guidance, its international submissions, its disclosure duties.
A country can be generous on paper and have built almost nothing, or cautious and highly organised, so the two are scored apart and reported as a pair. Each rests on three questions, scored 0, 0.5 or 1.
How far the law will protect intellectual property that a machine helped make.
How much administrative machinery sits behind that law.
The scale runs the same way on both axes: 0 is the open or developed end, 1 is the restrictive or absent one, and 0.5 sits between for anything conditional or untested. Read as a pair, the two scores show whether AI-assisted work can be protected here, and whether the office has said so in writing.
Within each axis the three questions carry equal weight. That is a deliberate choice, not an oversight. An AI developer might reasonably value freedom to train above patent inventorship, but weighting is where indices lose their credibility, so the score stays a flat mean and the choice is stated in the open. Each cell also cites a statute section, a case or an office page, so any single score can be challenged without disputing the rest.
Two positions are worth naming. A country can be open on the law but thin on machinery, an open-but-undeveloped stance that bets on ambiguity. Or it can be restrictive on rights yet run an active office, a restrictive-but-developed stance that reflects a settled policy choice rather than a gap. The pair of scores tells those two apart where a single ranking would blur them.
A rubric is only as good as the results it produces on cases where the answer is already known. These three jurisdictions sit outside this series and are scored with the same six questions, as a check. Their law is the best documented anywhere, so if the scale puts them where the law actually sits, the scale is doing its job. They are reference points, not editions.
Firm judicial settlement at both poles and the weakest statutory position on training. The courts have closed the door on AI inventors and AI authors, while everything on training rides on unresolved fair-use litigation. There is no per se duty to declare AI use in a filing, only a candour duty that bites where the use is material.
Thaler v. Vidal, 43 F.4th 1207 (Fed. Cir. 2022); Thaler v. Perlmutter (D.C. Cir. 2025), cert. denied 2 Mar 2026; Bartz v. Anthropic (N.D. Cal. 2025); USPTO revised inventorship guidance, 90 Fed. Reg. 54636 (28 Nov 2025); 89 Fed. Reg. 58128 (17 Jul 2024); 89 Fed. Reg. 25609 (11 Apr 2024).
The most permissive statutory position on training and the strongest transparency duty in force, sitting over an authorship question no court has answered. Read the label carefully: training and transparency run through EU law, but patents run through the EPO under the European Patent Convention, which is not an EU institution and covers 39 states.
Directive (EU) 2019/790, arts. 3 and 4; Regulation (EU) 2024/1689, art. 53(1)(c) and (d); EPO J 8/20 (2021); EPO Guidelines G-II 3.3.1 (2026 edition); Infopaq C-5/08; Painer C-145/10.
Liberal on patents since the Supreme Court abandoned the Aerotel test in February 2026, conservative on copyright, and currently thin on AI-specific machinery after the dedicated AI examination guidance was withdrawn. Section 9(3) nominally protects a work with no human author for 50 years, the widest AI-output right on paper, but no court has ever applied it to generative AI.
CDPA 1988 ss. 9(3), 12(7), 29A, 178; Thaler v Comptroller-General [2023] UKSC 49; Comptroller-General v Emotional Perception AI [2026] UKSC 3; UKIPO practice notice, 14 Jul 2026; Data (Use and Access) Act 2025 ss. 135 to 137.
The check earned its keep twice. The untested rule held: section 9(3) of the UK Copyright Act reads like the widest AI-output right anywhere, yet because no court has applied it to generative AI it scores as untested rather than open, which is the correct answer and the one a headline reading would miss. The check also exposed a limit worth stating. The Institution axis measures AI-specific machinery, not general office capacity, so a long-established office can score poorly when it has withdrawn its AI guidance and imposes no disclosure duty. Read that axis as what an office has built for AI, nothing wider.
Each row below carries the test we applied and the finding behind the score, with a source. The rationale, not the number, is the point.
Scores verified 20 July 2026, against AIP Index codebook v1.0. A score is only true as of its date.
What we look forA granted patent or a clearly workable examination route, not merely a silent statute.
The findingThe Patents Ordinance confines inventorship to the actual devisor of an invention, a human concept inherited from the United Kingdom statute, so an AI-assisted invention with a human inventor faces no bar while the machine itself cannot be named. No examination practice, guidance or grant evidence specific to AI-assisted inventions exists, and no DABUS-type filing has tested Pakistan.
What we look forProtection applied where human choices shape the work, and refusal of purely machine output.
The findingA human who uses AI as a tool and contributes authorship can hold copyright under the ordinary author definition, but there is no AI-specific rule and no registration practice to confirm it, so the position is untested. Purely AI-generated output has no route to protection, because the Copyright Ordinance of 1962 defines the author in exclusively human terms and, unlike the United Kingdom statute it derives from, carries no provision for a work with no human author.
What we look forA statutory or judicial mining lane. Silence scores against the miner, since the reserved-rights default governs.
The findingThere is no training or text-and-data-mining exception. The fair-dealing provision is a closed enumerated list, covering research, private study, criticism and a handful of other named purposes, with no open fair-use clause and no mining carve-out, and no reform is in motion. With copyright silent, the reserved-rights default governs the use of works for training.
What we look forA traceable, substantive submission on the record, not bare membership.
The findingPakistan is a WIPO member, but the WIPO Conversation submissions archive shows no traceable submission from Pakistan or its IP office, and no substantive session participation was found. Membership without a submission earns the middle.
What we look forA standalone AI examination text, not AI handled quietly under general practice.
The findingNo publicly available guidance from the IP office on examining AI-related filings was identified. The office's recent AI-related activity is an internal modernization of its own services, not examiner guidance for AI-originated filings. A reasonable-search negative.
What we look forA mandatory declaration inside the filing, not a proposal or an informal request.
The findingNo filing requires declaring AI use, and no in-force instrument imposes a labeling or transparency duty on AI-generated content. The cyber-crime code's false-information offence targets the content, not its provenance, and the National AI Policy's recommendation of future labeling rules has no enforceable text behind it.
Method. Six questions, three per axis, scored 0, 0.5 or 1 against AIP Index codebook v1.0. The three questions in each axis are weighted equally by choice, and the two axes are reported as a pair, never blended into one number. Every score cites a statute, case or office page, carries a verified date, and records its direction of travel apart from the level. Positions are editorial judgements read from primary law and office practice, not official scores. WIPO's public catalogue of IP-office AI initiatives does not score jurisdictions on this basis, so the readings here come straight from the source law. Not legal advice.
Pakistan governs artificial intelligence through general law and executive policy, not through a statute written for it. What binds is the cyber-crime code and its January 2025 amendment: a false-information offence and a new social-media regulator, none of it AI-specific. What points forward is the National AI Policy the federal cabinet approved in July 2025, a six-pillar roadmap with no enforceable duty behind it. The intellectual-property statutes predate the question: the Copyright Ordinance dates from 1962 and carries no provision for a work a machine helped make, so purely AI-generated output has no author and no protection. The live AI disputes here are criminal, deepfakes prosecuted under the cyber-crime code, rather than questions of ownership.
The site does three things: it maps every instrument that governs AI in Pakistan, 18 of them, ordered by how hard each one binds; it reads the same instruments a second time by how squarely each is written for AI; and it records the matters where AI meets the law, with what was searched and found empty stated as plainly as what was found. Every claim traces to a primary source, and absences are recorded rather than papered over.
A membership claim enters this table only from an official list: a declaration annex, a treaty signature table, an organization's own record or a government-to-government statement. Partner and observer statuses never become party status.
Know of a ruling or proceeding in Pakistan where AI is central? Send a source link; it will be reviewed before publishing.
Submit a case