Tanzania

Case management

In 2022, the Tanzanian judiciary entered into a four-year, approximately USD $3 million agreement with Almawave, an Italian AI company, to implement AI-driven transcription and translation services in courtrooms. This initiative leverages ‘Pervoice’, Almawave’s proprietary speech-recognition technology, to automate the recording and translation of court proceedings between Kiswahili and English, helping to address the country’s linguistic diversity and the operational burden of manual transcription. The system is trained on a range of Kiswahili dialects spoken across mainland Tanzania and the islands to improve accuracy and support real-time or near-real-time use in legal settings. The AI system has been installed in 11 courtrooms, with plans to expand to around 50 of Tanzania’s 169 courtrooms over the lifetime of the contract.

Law Companion (discussed below) is an AI-powered legal assistant developed in Nigeria, used by court staff in Tanzania. The software allows court users, including judges, to capture audio and video recordings of hearings, and offers automatic transcription within a matter of seconds. The system can also translate transcripts from Kiswahili to English.

Legal research, analysis and drafting support

Law Companion is an AI-powered legal assistant developed in Nigeria, used by court staff in Tanzania. The software offers case analysis, drafting support, and the ability to be ‘fed caselaw and statutes for referencing’. As at July 2026, the tool is in use in the High Court of Tanzania and Court of Appeal. Specifically, the system can:

1. Generate a first draft of judgments or draft documents from a transcript of proceedings, or from uploaded case materials. Judges must still make final edits and sign the judgment.
2. Analyse a written judgment and suggest possible grounds of appeal or legal gaps, flagging potential weaknesses.
3. Check whether a case actually exists when the tool is fed appropriate case databases, such as Tanzanian law.
4. Open windows that show related cases and laws when drafting.
5. Provide predictive flags for case-ageing and backlog, and reminders about time-bar issues.

The system has a reported 98% accuracy rate and can be trained on different Kiswahili dialects.

Under the pilot TTS programme launched in 2024 across 11 judicial zones, judges can dictate rulings which are then transcribed and translated. Previously, writing a 200-page judgment could take weeks. The introduction of AI-supported transcription has enhanced efficiency, reduced paperwork, and mitigated physical strain for judicial officers. Human oversight is maintained: the AI system's outputs can be reviewed by human court staff before becoming official documentation.

Decision-making support

Law Companion (discussed above), is an AI software that can generate a first draft of judgments. Judges must still make final edits and sign the judgment. As at July 2026, this software is in use in the High Court of Tanzania and Court of Appeal.

Defence

Legal research, analysis and drafting support

Lawyers in Tanzania have informally experimented with general AI-powered tools (ChatGPT, Bing Chat, and legal research platforms with AI assistance) to conduct legal research, summarise caselaw, draft pleadings, assist with contract analysis, and generate draft documents.

Victims

Under Tanzanian criminal procedure, victims do not have direct party standing in criminal proceedings: prosecutions are conducted by the Director of Public Prosecutions (DPP) under Article 59B of the Constitution of the United Republic of Tanzania (1977). However, victims have indirect standing and participatory rights: (a) under section 129 of the Criminal Procedure Act (Cap 20 R.E. 2022), a victim or complainant may institute a complaint that leads to a criminal charge; (b) under section 368 of the Criminal Procedure Act, courts may order a convicted person to pay compensation to any person who has suffered material loss or personal injury in consequence of the offence; and (c) under section 145 of the Criminal Procedure Act, the court may issue a summons to any person to attend as a witness, including victims.

TRAINING

As at July 2026, only limited training is available to legal practitioners in Tanzania, and it is not yet systematic nor mandatory. However, Tanzania’s Draft Guidelines for the Secure and Ethical Use of AI (discussed below) provide that ‘[e]ducating entity/institution employees on the ethical implications of AI and providing them with training on the use of ethical AI is essential’, to assist employees in understanding how to identify and mitigate potential ethical issues.

For judges and lawyers, reported training initiatives include:

1. Individual judges have been reported to deliver practical AI training to judicial officers and wider legal stakeholders (including magistrates and lawyers), covering use of software such as ChatGPT, CoPilot and other AI tools for legal research, drafting, and analysis in the judiciary.
2. In a broader sense, the Tanzania Public Service College, in partnership with the judiciary, ran seven-week training programmes in 2024 on digital skills for more than 1,800 judiciary staff, including training on digital tools and e-judiciary system.
3. It has been reported that the Tanzanian judiciary is running over 17 ICT-focused programmes covering system upgrades, staff training, and personnel development.

At the regional level, UNESCO and the Smart Africa Digital Academy have conducted interregional training programmes to promote responsible AI adoption in the legal system across African countries, including Tanzania.

UNESCO has supported the development of a competency-based Judicial Curriculum on Artificial Intelligence and the Rule of Law for the Institute of Judicial Administration (IJA) in Lushoto, Tanzania. It is grounded in five UNESCO normative instruments: the Recommendation on the Ethics of Artificial Intelligence (2021), the Global Toolkit on AI and the Rule of Law for the Judiciary, the Guidelines for the Use of AI Systems in Courts and Tribunals (2025), the Ethical Impact Assessment methodology, and the Guidance for Generative AI in Education and Research (2023). The curriculum combines real-life case studies, judicial simulations and practical exercises. UNESCO has also developed a comprehensive instructor’s guide for IJA faculty with lesson plans, exercises and assessment tools. The initiative is funded by the European Union.

The Lawyers Hub Africa has recommended that judiciaries in Africa should establish an Artificial Intelligence Capacity Building Initiative to enhance AI literacy and awareness by organising Training of Trainers (ToT) courses, integrating educational materials into curricula, and fostering skills development in AI use and implementation. Specialised training programmes for judges, legal professionals, and AI developers should cover topics such as gender issues, AI ethics, and the implications of AI in judicial decision-making, including understanding and mitigating biases that can manifest in AI algorithms.

Universities are also expanding AI-related degree programmes: the University of Dar es Salaam (UDSM), Nelson Mandela African Institution of Science and Technology (NM-AIST), Mbeya University of Science & Technology (MUST), and IIT-Madras Zanzibar Campus offer specialised AI and Data Science degrees. The National Guidelines for Artificial Intelligence in Education (January 2025) (discussed below) mandate continuous professional development programmes on AI literacy and ethical use for educators, administrators, and learners at all education levels.

In July 2026, UNESCO Dar es Salaam partnered with the Zanzibar Correctional Facility (ZCF) to deliver a three-day Capacity-Building Programme on 21st Century Skills, Digital Pedagogy, Practical Digital Tools and Artificial Intelligence (14–16 July 2026). The programme trained 32 correctional officers from departments including Administration and Management, Rehabilitation and Programmes, ICT, Custody and Operations, and the ZCF Training College. Sessions covered prompt engineering, AI-assisted research, responsible AI use, detection of AI-generated and manipulated media, knowledge management using AI, and offline AI solutions for secure institutional environments. This represents the first publicly reported AI training initiative for a Tanzanian criminal justice institution outside the judiciary.

REGULATION

Tanzania has no dedicated AI legislation and no designated regulatory body overseeing AI use, with responsibility instead distributed across sector-specific institutions and a growing but still-developing set of policy instruments, strategies, and draft guidelines. As at July 2026, the government's approach rests primarily on the Draft Guidelines for the Secure and Ethical Use of Artificial Intelligence and a draft National AI Strategy, both non-binding, alongside a set of digital and sector-specific policies that acknowledge AI without directly regulating it. There are no judicial or Bar Council guidelines for legal practitioners on the use of AI, nor any Criminal Procedure Rules that specifically address AI in criminal proceedings. Instead, existing general laws — including the Evidence Act, the Electronic Transactions Act, the Personal Data Protection Act 2022, the Cybercrimes Act 2015, and constitutional rights protections — may be construed to apply to AI use in criminal and court proceedings, though significant gaps remain around liability, transparency, and the admissibility of AI-generated evidence.

AI regulations

As at July 2026, there is no AI-specific binding legislation in Tanzania. Despite increasing adoption of AI, Tanzania does not have a designated regulatory body overseeing its use, with AI applications instead falling under multiple institutions depending on their sector of deployment: the Tanzania Communications Regulatory Authority may regulate AI used in communication technologies; the Ministry of Education may oversee AI used in educational tools; and the Ministry of Agriculture may address AI-driven innovations in precision farming.

Draft Guidelines for the Secure and Ethical Use of Artificial Intelligence (2024)

The 2024 Draft Guidelines for the Secure and Ethical Use of Artificial Intelligence establish a national framework for developing and deploying AI in ways that support economic growth while protecting human rights, privacy, fairness, and public trust. The Guidelines emphasise that AI systems should be human-centred, transparent, accountable, inclusive, and secure, with safeguards against bias, discrimination, misinformation, cybercrime, and misuse of personal data.

The Guidelines adopt an AI Code of Ethics for AI developers to adopt, recommending that the AI system:

1. Is always aligned with the entity/institution mission and values;
2. Strives for excellence and quality;
3. Adheres to ethical principles;
4. Assists the entity/institution in unleashing human potential easily and affordably;
5. Assists the entity/institution in unleashing human potential by generating value and positive impact;
6. Is open and transparent;
7. Is fair, inclusive, and diverse;
8. Be reproducible;
9. Has a positive impact on people;
10. Does not discriminate in any way against individuals and/or entity/institution;
11. Be in line with the use of AI models that allow commercial use and free use for research purposes;
12. Promotes ethical conduct in data pipelines and the use of models, from data collection for model training; training, evaluation, and inspection of results; and deployment for use in development and production environments.

The Guidelines aim to assist in ‘managing risks for a trustworthy ethical AI’, and provide as follows:

The Tanzania Commission for Science and Technology is responsible for monitoring and coordinating all activities relating to scientific research, technology development and innovation. In the context of AI, the Commission oversees research ethics on the Mainland through its National Research Clearance Committee, and has developed a National Data Sharing Framework for Science, Technology, and Innovation. The Commission supports AI-related research, funding, and innovation — including through the Buni Hub in Dar es Salaam, which focuses on innovation and tech entrepreneurship. However, the Commission’s existing research ethics guidelines predate the widespread proliferation of AI and may not adequately address the unique ethical challenges it poses.

Other AI-specific policies and frameworks

Other Tanzanian policies and strategies relevant to AI, including sector-specific developments, include (as set out in the UNESCO AI Readiness Report):

Guidelines for practitioners

As at July 2026, Tanzania does not have any judicial or Bar Council, or professional regulatory guidance on the use of AI by legal practitioners. Tanzanian judicial officers are engaging with international bodies, such as UNESCO, regarding guidelines for using AI in courts, ensuring ethical standards and human oversight.

UNESCO has supported the development of a competency-based judicial curriculum on AI and the Rule of Law at the Institute of Judicial Administration in Lushoto, which is grounded in several UNESCO normative instruments including the Recommendation on the Ethics of Artificial Intelligence and the Guidelines for the Use of AI Systems in Courts and Tribunals (see ‘Training’ section).

Criminal procedure rules

The use of AI, and especially the admissibility of AI-generated evidence, in Tanzanian criminal proceedings is governed by existing laws that do not explicitly address AI.

The Evidence Act (2023)

The Evidence Act (Cap 6 R.E. 2023) establishes foundational principles governing the admissibility and relevance of evidence, including AI evidence:

• Under Section 7, evidence is relevant if it makes a fact in issue more or less probable, meaning AI evidence must be shown to directly relate to the facts of the case.
• Section 8 stipulates that for evidence to be admissible it must be relevant, and courts may require expert testimony to establish the reliability of AI processes used.
• AI-generated reports or data can be classified as documentary evidence under Section 3 read together with Section 67 of the Act, which outlines how such documents must be authentic and reliable to be admissible.

Electronic Transactions Act (2015)

The Electronic Transactions Act (Cap 442 R.E. 2015) provides a legal framework for electronic records and communications relevant to AI evidence.

• Section 3 recognises electronic records as having legal validity, which is crucial for AI-generated evidence that exists in digital form. Section 10 outlines conditions under which electronic records can be admissible in court, stating that such records are admissible if they are relevant and have not been tampered with.
• Section 11 emphasises the importance of maintaining the integrity of electronic records, meaning AI evidence must be shown to be authentic and reliable.
• Section 12 creates a presumption that electronic records are accurate unless proven otherwise, which can support the relevance of AI evidence provided it adheres to legal standards.

Data protection legislation

Personal Data Protection Act, Chapter 44 (Revised Edition of 2022)

The Personal Data Protection Act (PDPA) enacted by Parliament in November 2022 and brought into force on 1 May 2023, is the cornerstone of data governance in Tanzania. The Act applies throughout Mainland Tanzania and Zanzibar (with application in Zanzibar limited to Union matters).

Key provisions relevant to AI in criminal proceedings include:

1. Broad definition of 'personal data' encompassing information relating to an identified or identifiable natural person, including identifiers and characteristics specific to physical, physiological, genetic, mental, economic, cultural, or social identity;
2. Special protections for 'sensitive personal data' (covering race, ethnicity, religion, health, biometrics, criminal records) with processing generally prohibited subject to specific exceptions;
3. Core principles under Section 5 requiring data to be processed lawfully, fairly, and transparently, collected for explicit and legitimate purposes, and subject to data minimisation and storage limitation requirements;
4. Data subject rights including the right to be informed about processing, access data, correct inaccuracies, object to processing (particularly for direct marketing), and rights related to automated decision-making; and
5. Enforcement mechanisms through the Personal Data Protection Commission (PDPC), established as an independent regulatory authority operational since May 2023, responsible for registering data collectors and processors, investigating complaints, and imposing fines up to TZS 20 million and imprisonment up to 10 years for certain offences.

The PDPA is supplemented by the Personal Data Protection (Personal Data Collection and Processing) Regulations, 2023. Key provisions relevant to AI include: regulation 19(3), which requires entities to inform data subjects where personal data is processed through automated decision-making, including AI systems; regulations 23 to 30, which elaborate the fundamental data protection principles (lawfulness, fairness, transparency, purpose limitation, data minimisation, accuracy, storage limitation, integrity and confidentiality, and accountability); regulation 25, which requires a lawful basis for processing; and regulations 20 to 22, which govern cross-border data transfers — a key consideration for AI systems relying on cloud-based or overseas processing. The PDPC also issued guidance templates in March 2025 for data breach notifications, Data Protection Impact Assessments, and data sharing agreements.

Cybersecurity laws

Cybercrimes Act 2015 (Act Number 14 of 2015)

The Cybercrimes Act 2015 (Act No. 14 of 2015) provides the primary legal framework for addressing offences related to computer systems and data. The Act establishes the following offences: illegal access (hacking) (section 4); illegal remaining in a computer system (section 5); illegal interception of data/communications (section 6); data interference (unlawful destruction, alteration, or deletion) (section 7); data espionage (unauthorised obtaining of protected data) (section 8); illegal system interference (section 9); illegal devices (e.g. malware) (section 10); computer-related forgery (section 11); computer-related fraud (section 12); publication of false information (section 16) and publication of racist or xenophobic material (section 17). Investigation powers are in Part IV (sections 31-38). The Act grants specific powers to law enforcement officers (including police, intelligence services, prosecutors, and other authorised personnel) for investigation, search, and seizure.

While it does not specifically address AI, any AI system used to commit cybercrimes would fall under its provisions. The Act is directly relevant to protecting AI systems from malicious attacks (e.g. unauthorised access, data poisoning, model tampering) and prosecuting the misuse of AI for criminal activities like fraud.

The Act was under review in 2024 and may be amended to incorporate AI-specific provisions. Tanzania’s Government Cyber Security Strategy 2022–2027, published by the President's Office, Public Service Management and Good Governance (PO-PSMGG), sets out the framework for cybersecurity in the public sector. The Strategy is built around five objectives: (i) improving research and development in cyber security technologies; (ii) enhancing the protection of government ICT infrastructure and e-services; (iii) strengthening the cyber security and resilience of ICT systems; (iv) improving cyber security skills, knowledge and awareness; and (v) enhancing the management of cyber security risks and incidents. The Strategy includes specific targets such as connecting 660 public institution sites to the Government Network (GovNet) by June 2027, training 500 ICT security personnel, and developing incident response management plans across all public institutions.

Human rights

Fundamental rights related to due process and accountability are enshrined in the Constitution of the United Republic of Tanzania, which serves as the supreme law. Tanzania's Constitution guarantees privacy, protection from arbitrary interference, and the right to information. These fundamental rights may come into play when AI systems process personal or sensitive information. The constitutional guarantees underpin the legal framework for AI governance, including:

• Right to equality (Article 12) — equality before the law and equal protection, relevant to algorithmic bias;
• Right to privacy (Article 16) — protection from arbitrary interference with privacy, correspondence, and communications;
• Right to a fair trial (Article 13(6)(a)) — right to a fair hearing by an impartial court or tribunal, relevant where AI tools influence judicial decision-making;
• Right to freedom of expression and information (Article 18) — right to seek, receive, and impart information;
• Right to life, personal freedom, and security (Articles 14-15) — protection from arbitrary deprivation of liberty;
• Right to be heard (Article 13(6)(a)) — the right to be heard before a decision affecting one’s rights.

Tanzania ratified the International Covenant on Civil and Political Rights (ICCPR) in 1976 and the United Nations Convention on the Rights of the Child (CRC) in 1991. Relevant ICCPR provisions include Article 2 (non-discrimination); Article 9 (liberty and security of person); Article 14 (fair trial and equality before courts); Article 17 (privacy); and Article 26 (equality before the law). Relevant CRC provisions include Article 2 (non-discrimination); Article 3 (best interests of the child); Article 12 (right to be heard); Article 16 (privacy); Article 37 (protection from torture, cruel treatment, and deprivation of liberty); and Article 40 (juvenile justice guarantees). These international obligations inform the standards applicable to AI use in criminal proceedings involving children and vulnerable persons.

Outlook

As at July 2026, the Ministry of Communication and Information Technology is leading development of a draft National AI Strategy and Guidelines, with a clear focus on aligning with major global and regional frameworks including the UNESCO Recommendation on the Ethics of Artificial Intelligence and the African Union Continental Strategy on AI (adopted in 2024). The draft National AI Strategy was released in July 2025 and, as at July 2026, remains in draft form and has not been formally adopted as binding policy.

The Tanzania Digital Economy Strategic Framework 2024-2034 articulates a vision to 'transform Tanzanians through cutting-edge, supportive, and affordable digital technologies, solutions and skills in all areas of socio-economic activities.' Tanzania Development Vision 2050 also aspires to transform the country into a technology-driven, high-productivity economy, emphasising the urgent need to develop and apply emerging technologies including AI.

CASES

As at July 2026, there is no reported case law specifically addressing the use of AI in legal proceedings in Tanzania.