Zambia

Defence

Legal research, analysis and drafting support

Mulawo is an AI-enabled legal research and document-analysis platform focused on Zambian law. According to its website, users can search legislation, statutory instruments, judgments, constitutional documents, Gazettes and other official legal publications using natural-language queries. The platform presents its legal research answers with citations and links to the underlying sources and also offers functionality for analysing users’ own documents. Mulawo states that it provides legal information rather than legal advice. Publicly available information does not establish the completeness of its legal corpus, the accuracy of its outputs, or the extent to which it is used by legal practitioners in criminal proceedings.

Similarly, Lamulo.io is an AI-powered legal research platform designed for the Zambian legal environment. According to its website, its legal corpus includes Zambia’s Constitution, statutes, case law and customary law. The platform offers legal research functionality including AI-assisted questions and answers with citations to Zambian legal authorities, statutory analysis and case-precedent search, as well as multilingual support for English and major Zambian languages.

Informal reports also suggest legal firms in Zambia are beginning to adopt tools such as ChatGPT in their everyday tasks.

Victims

Under Zambian law, criminal prosecutions are conducted by the Director of Public Prosecutions and public prosecutors, with victims participating primarily as complainants or witnesses. However, Article 180(8) of the Zambian Constitution in connection with Section 90 of the Criminal Procedure Code does recognise very limited mechanisms for private prosecutions subject to the consent of the Director of Public Prosecutions (cf. Wang Shunxue v Attorney General and Wang Qinghai, Case No. 2021/CCZ/003 (17 September 2021)), while section 30 of the Penal Code empowers the court to award compensation following conviction.

As at July 2026, there are no publicly reported examples of victims in Zambia using AI tools in criminal proceedings.

TRAINING

There is no mandatory or systematic training on AI for judges, prosecutors, law enforcement, or legal practitioners in Zambia as at July 2026. However, the government's strategies point to future capacity-building: The National Digital Transformation Strategy 2023–2027 identifies digital literacy and skills as a core pillar and includes targeted measures to expand Information and Communication Technology (‘ICT’) expertise, including training 5,000 ICT professionals in new and emerging technologies such as artificial intelligence, robotics, machine learning, blockchain and big data.

In a related initiative, though not focused on AI, the Zambia Information and Communications Technology Authority, partnering with the International Telecommunications Union (a UN organisation), launched a two-day in-service training programme for prosecutors on 19 November 2024 in Lusaka, which focused on cybersecurity and electronic evidence handling. A total of 50 Zambian prosecutors participated in this training, with the expectation of being able to handle the increasing threat of cybercrime.

REGULATION

As at July 2026, Zambia has no dedicated legislation specifically regulating the use of AI in criminal proceedings or in court proceedings more generally. The country's regulatory approach to AI remains in development, and relies on several existing laws and regulatory frameworks that may apply to the use of AI tools in criminal proceedings, despite not specifically mentioning AI.

AI regulations

There is no AI-specific legislation in force in Zambia as at July 2026.

The National Artificial Intelligence Strategy 2024-2026, developed by the Ministry of Technology and Science, sets out a policy framework for AI development and adoption across sectors, including references to ethical principles and governance, but it does not establish binding legal requirements for AI use in the justice sector. The Strategy prioritises human capital development, digital infrastructure, data ecosystems, and sectoral AI adoption, and establishes governance and ethical oversight through a proposed National AI Council. It does not create binding legal requirements for AI use in the justice sector.

Guidelines for practitioners

As at July 2026, there are no AI-specific guidelines for practitioners involved in criminal proceedings in Zambia. There are, however, several existing professional ethics frameworks that may apply to the use of AI in criminal proceedings.

Judicial Code of Conduct, 1999

The Judicial (Code of Conduct) Act 1999 establishes ethical standards for judicial officers in Zambia. Section 2 of the Act defines ‘judicial officer’ broadly to include the Chief Justice, Deputy Chief Justice, judges, the Chairman and Deputy Chairman of the Industrial Relations Court, registrars, magistrates, and justices of local courts or other persons having power to hold or exercise the judicial powers of a court.

The Act does not specifically address AI, but its provisions would apply to any judicial use of AI tools. Key provisions relevant to use of AI include:

• Independence and impartiality (Section 3): A judicial officer shall uphold the integrity, independence and impartiality of the judicature in accordance with the Constitution, the Act, or any other law.
• Non-discrimination (Section 4): A judicial officer shall perform duties without bias or prejudice and shall not manifest discrimination based upon race, tribe, sex, place of origin, marital status, political opinion, colour or creed.

These provisions would require judges using AI tools to maintain personal responsibility for their decisions, ensure that AI assistance does not introduce bias or compromise impartiality, and refrain from delegating the core judicial function to automated systems.

Ethical Guidelines for Lawyers in Zambia

There are two regulatory frameworks that may impose obligations on Zambian lawyers using AI in criminal proceedings:

1. The Legal Practitioners Act (commenced on 23 March 1973, with the latest edition dating 24 March 2014) which governs the legal profession in Zambia, including admission, enrolment, practising certificates, and professional misconduct;
2. The Law Association of Zambia Act 1973, which establishes the Law Association of Zambia, a professional body that exercises regulatory functions for legal practitioners, including issuing practising certificates.

Neither Act specifically addresses AI, but they establish general professional standards that would apply to lawyers using AI tools. In particular, Part VII of the Legal Practitioners Act addresses privileges, restrictions, and offences in connection with practice. These standards would require lawyers to verify any AI-generated legal research or submissions for accuracy, including checking for hallucinations such as fabricated case citations or legal authorities.

The Law Association of Zambia has not yet issued guidance specifically addressing the use of AI tools by legal practitioners in Zambia. Careless use of AI tools, such as submitting hallucinated case citations, could nonetheless expose a practitioner to disciplinary proceedings under the Legal Practitioners Act and the Legal Practitioners (Disciplinary Proceedings) Rules referred to in Part IV and Section 23 of the Legal Practitioners Act.

Criminal procedure rules

Zambia’s criminal procedure laws do not specifically mention AI or algorithmic evidence, but they contain general provisions that would apply to technology in the courtroom.

Criminal Procedure Code, 1931

The Criminal Procedure Code 1931 governs how criminal trials are conducted. While it does not specifically address AI, its provisions on evidence admissibility, the rights of the accused, and procedural fairness would apply to AI-generated or AI-analysed evidence.

Relevant provisions of the Criminal Procedure Code include, in particular, Section 191, which generally requires evidence to be taken in the presence of the accused or their advocate; Sections 191A and 192 which, inter alia, deal with expert witnesses in connection with medical reports and instances where a fact is ascertained by a process, Section 286, concerning notice of additional prosecution witnesses, and Section 287, which permits the accused or their advocate to cross-examine prosecution witnesses. These provisions may enable an accused to examine and challenge evidence produced or analysed using AI. However, the Code does not expressly regulate AI, require disclosure of AI involvement, or establish specific requirements concerning AI explainability or bias.

Evidence Act, 1967

Zambia's Evidence Act governs documentary evidence in criminal trials. The Evidence Act provides for the admissibility of documentary evidence as to facts in issue and includes specific provisions for certain trade, business, or professional records in criminal proceedings. In particular, Section 2 defines a ‘document’ broadly to include any device by means of which information is recorded or stored. Section 3 governs documentary evidence as to facts in issue in civil proceedings, while Section 4 provides for the admissibility in criminal proceedings of records relating to a trade, business or profession, subject to specified conditions concerning the source and recording of the information. Under Section 5, when assessing the weight of documentary evidence admitted under the Act, the court must consider all circumstances affecting its accuracy, including whether the statement was made contemporaneously with the relevant facts and whether the supplier or recorder of the information had any incentive to conceal or misrepresent those facts. While the Act does not specifically address artificial intelligence or modern digital forensics, these principles could also be relevant when assessing AI-generated or AI-assisted evidence.

Electronic Communications and Transactions Act, 2021

The Electronic Communications and Transactions Act 2021 establishes a framework for electronic evidence communications, data messages, and transactions, aiming to promote secure electronic signatures, facilitate electronic filing of documents by public authorities; provide for the use, security, facilitation and regulation of electronic communications and transactions, promote legal certainty and confidence, and encourage investment and innovation in relation to electronic transactions.

While not AI-specific, the Act's provisions on the admissibility and evidential weight of electronic records would apply to AI outputs used as evidence in criminal proceedings. In particular, Section 8 sets out the criteria for determining the originality and integrity of a data message and Section 9(2) of the Act provides that information in the form of a data message is to be given due evidential weight, while Section 9(3) requires the court, when assessing that evidential weight, to have regard to the reliability of the manner in which the data message was generated, stored or communicated, the reliability of the manner in which its integrity was maintained, the manner in which its originator was identified, and any other relevant factor. The Act also includes provisions on automated transactions and the use of electronic agents (Section 14), which could encompass AI systems operating without direct human intervention. This interpretation is, however, subject to how Zambian courts and practitioners construe and apply these provisions in practice.

Penal Code, 1930

Section 108 (Fabricating evidence) of Zambia's Penal Code makes it a misdemeanour, with intent to mislead any tribunal in any judicial proceeding, to fabricate evidence by any means other than perjury, or to knowingly make use of such fabricated evidence; the prescribed maximum penalty is imprisonment for seven years. This provision may be relevant to intentionally manipulating digital evidence, including creating or deploying deepfakes (AI-generated synthetic media) with intent to mislead a court.

Data protection legislation

Data Protection Act, 2021

The Data Protection Act 2021, establishes a comprehensive framework for the protection of personal data in Zambia. The Act applies to the processing of personal data by both public and private entities, including law enforcement agencies, courts, and legal practitioners. The Act requires that personal data be processed lawfully and transparently and sets conditions for data sharing and security.

The Data Protection Act does not specifically mention AI, but its Part IX grants a number of rights to data subjects which would also apply to AI use. In particular, Section 62 addresses automated decision-making: individuals have the right not to be subject to purely automated decisions that have significant effects on them, unless certain safeguards exist (such as a legal authorisation or explicit consent). This is relevant for any future use of AI in profiling or decision support during investigations or sentencing. For example, if an AI system were introduced to recommend whether a suspect should be granted bail, Section 62 suggests that a person could object to a bail decision made entirely by a machine.

The Data Protection Act also establishes the role of the Data Protection Commissioner under Section 4 and provides for the appointment and functions of the Data Protection Commissioner under Sections 5 and 6. The Office is responsible for regulating data protection and privacy in Zambia, registering data controllers and processors, licensing data auditors, investigating complaints, advising the Government on data protection matters, and promoting compliance with the Act. This includes the power to compound certain offenses (Section 74) and to issue guidelines (Section 79). The Data Protection Commissioner could therefore issue directives affecting how police and courts handle personal data in AI tools.

In April 2025 the Zambian Cabinet formally approved in principle the introduction of a Bill to repeal and replace the Data Protection Act No. 3 of 2021 with a new law that would expand the legal framework to cover emerging technologies such as AI, machine learning, and data analytics, as well as broader categories of data including non-personal, industrial, and public data, aligning with global trends in data governance. If enacted, this new law could provide a more explicit statutory basis for regulating AI systems, potentially addressing issues like data bias, algorithmic transparency, and accountability—all of which are crucial for criminal justice applications.

Cybersecurity laws

Zambia’s cybersecurity legislation chiefly regulates crimes and security measures rather than the use of AI in court, but since the provisions are largely technology-neutral, some of them will also be relevant to AI.

Cyber Security Act, 2025

The Cyber Security Act 2025 contains Zambia's framework for cybersecurity. The Act designates certain systems as critical information infrastructure, which may include justice sector systems. Enhanced security requirements apply to such systems, which would extend to any AI components. Although the Act does not expressly refer to AI, its broadly framed and technology-neutral definitions of ‘computer systems’, ‘computer data’ and ‘cyber attacks’ may be sufficiently wide to encompass AI systems and AI-driven processes.

According to Section 37(1) of the Cyber Security Act, 2025, ‘a person shall not use an electronic communications service, critical information or critical information infrastructure to randomly monitor a communication, except for mechanical or service quality control checks.’ This provision could apply to AI systems used for automated monitoring or surveillance, as the prohibition extends to any form of monitoring (defined to include listening to, viewing, reading or recording communications) carried out through electronic, software or other devices.

Moreover, the Act imposes concrete obligations on controllers of critical information and critical information infrastructure, including the duty to conduct cyber audits (Section 14), to submit cyber security situational awareness reports (Section 16), and to notify and report cyber security incidents (Section 17), alongside the maintenance of a Cyber Security Risk Register (Section 20). These provisions require the identification, assessment and mitigation of risks affecting computer systems and data, regardless of their technological origin, meaning that risks arising from AI systems (such as automated processing, large-scale data analysis, or system vulnerabilities) would need to be addressed within these existing statutory compliance mechanisms.

The Cyber Security Act establishes the Zambia Cyber Incident Response Team under Section 6, with functions including incident response, coordination, risk assessment and the issuance of alerts, supported by sectoral response teams under Section 7 and reporting obligations under Section 17.

Cyber Crimes Act, 2025

The Cyber Crimes Act 2025, while not specifically mentioning AI, criminalises a range of conduct relevant to AI misuse and to the integrity of electronic evidence. For example:

Read together, these provisions indicate that AI-enabled misconduct is already captured through the Act's focus on the manipulation, transmission and impact of computer data, rather than through technology-specific regulation.

The enactment of the Cyber Security Act, 2025 and the Cyber Crimes Act, 2025 has attracted mixed reactions from stakeholders. The Global Network Initiative (GNI) issued a formal statement in July 2025 expressing serious concern that both cybersecurity Acts were developed with minimal transparency and public participation, and that their substance raises significant risks to freedom of expression, access to information, and privacy. A technology expert commentary in August 2025 acknowledged the laws as a ‘watershed moment in the era of AI’ but cautioned that translating policy into genuine capacity and responsible adoption remains the key challenge.

Human rights

Constitution of Zambia, 1999

The Constitution of Zambia 1991 contains a number of relevant provisions to the use of AI in criminal proceedings:

International and regional human rights instruments

Zambia is a party to several international human rights instruments that may provide additional guidance on AI use in criminal proceedings:

Outlook

Zambia's approach to AI in the justice sector is likely to evolve alongside its broader digital transformation agenda. The National Artificial Intelligence Strategy 2024–2026 implementation runs in three phases through 2026—foundational (first 100 days), development (year one), and scaling (year two)—but as at July 2026, no draft AI bill has followed the National Assembly's June 2025 motion urging a dedicated legal framework. Cabinet's April 2025 in-principle approval of a Bill to repeal and replace the Data Protection Act 2021 suggests Zambia's first statutory anchor for AI-relevant data governance may emerge during the next legislative cycle, running alongside the Data Protection Commissioner's first enforcement cycle under the existing Act.

The Judiciary's digitalisation programme continues through its partnership with Synergy International Systems, and the CIJ/Adalat AI partnership (announced October 2025) includes Zambia among its target countries for future AI-driven justice solutions. In the absence of AI-specific legislation, courts are likely to apply existing constitutional principles—including judicial independence and fair trial guarantees—to emerging issues such as the reliability of AI-generated evidence, drawing on approaches seen in other common law jurisdictions.

A CIVICUS Report (April 2026) notes that none of the 11 African countries studied—including Zambia—has, as at July 2026, a legal framework capable of balancing security needs against human rights commitments, and questions the security rationale behind Zambia's substantial ‘Safe City’ surveillance spending given its comparatively low terrorism threat. At the international level, Zambia participates in UNESCO's AI Readiness Assessment and its Africa AI-and-justice capacity-building track, but has not formally adopted the UNESCO Guidelines for the Use of AI Systems in Courts and Tribunals (2025) as at July 2026.

CASES

Cybersecurity

In a 2023 Lusaka prosecution, offenders reportedly used AI-enabled voice cloning to impersonate individuals and induce victims to transfer funds. The matter was prosecuted under the Cyber Security and Cyber Crimes Acts, which addressed computer-related misrepresentation, with the court treating the conduct as fraud and digital impersonation rather than developing any AI-specific legal principles.

Similarly, in a 2024 Lusaka case reported by the National Prosecution Authority, more than twenty people were convicted for operating a large-scale cybercrime scheme involving fake identities, scripted communications, and coordinated digital deception. Although that case was not framed in terms of AI, the conduct displayed features analogous to AI-enabled fraud – including scalability, identity simulation, and automated interaction – and was similarly prosecuted under technology-neutral offences such as computer-related misrepresentation and identity-related crimes.