Table of Contents
Algeria's national AI strategy is not yet a finished rulebook. A first version was presented in December 2024, the government reviewed a draft in May 2026, and implementation meetings began in August 2026. For industry, it signals more national computing capacity, more AI graduates and a push to keep data in the country, but no new obligations yet.
Key takeaways
- The strategy has moved in three public steps: a first version in December 2024, a government review of a draft in May 2026, and a first implementation meeting on 5 August 2026.
- The May 2026 draft rests on three levers: data, digital infrastructure and human skills. It was still to be adopted by the Council of Ministers (Radio Algérie).
- Industry was named a priority sector in December 2024, alongside agriculture and health (Radio Algérie).
- There is no AI-specific law that we know of. The personal data law and the public procurement law already shape AI projects.
- Companies do not need to wait: classify data, keep sensitive data on site, build skills and use an existing governance framework.
How did Algeria's national AI strategy come about?
The strategy has been built in public steps, each reported by official media. The table summarises them.
| Date | Step | What the official source says |
|---|---|---|
| 7 December 2024 | First version presented at the 3rd African Startup Conference in Algiers | Presented by Merouane Debbah, president of the Scientific Council for AI, which was set up to draft this first version. Axes include research, skills, start-up support and data centres (Radio Algérie) |
| 25 May 2026 | Government meeting chaired by the Prime Minister reviews the draft strategy | Three levers: data, digital infrastructure and human skills. Adoption by the Council of Ministers, then action plans (Radio Algérie) |
| 5 August 2026 | First coordination meeting on implementation, chaired by the Minister of Higher Education and Scientific Research | Attended by several ministries and national bodies (APS); an interministerial monitoring committee was created (Dzair Tube) |
In December 2024, APS reported that the strategy was "adopted" at the conference's closing ceremony. Debbah said it had six main axes and that it would reduce dependence on solutions imported from abroad. The Scientific Council is described as an advisory body supervised by the ministries of higher education and of the knowledge economy and start-ups (Radio Algérie).
Is the strategy final, and what is still a draft?
The wording of the sources matters here. The May 2026 government release calls the text a "draft" and says it will become action plans "following its adoption by the Council of Ministers" (Radio Algérie). On 5 August 2026, the Algerian technology magazine IT Mag described the strategy as submitted to the Council of Ministers for adoption in 2026, with a vision to 2035 (IT Mag).
At the time of writing, we have not found a published full text of the strategy or a Council of Ministers release announcing its adoption. Implementation has nonetheless started.
What is not public yet:
- The full text and the final list of axes. The December 2024 report names only some of the six axes.
- The budget. In December 2024, Debbah said the authorities had set aside "a special budget", without giving an amount (Radio Algérie).
- The action plans and deadlines. The August 2026 committee works "according to a defined timetable", which has not been published (Dzair Tube).
- Any binding rules for companies. Nothing published so far creates obligations for industrial firms.
Figures you may see quoted in the press, such as market sizes or training targets, should be traced to an official source before you put them in a business case.
What does the strategy prioritise?
The priorities announced at the August 2026 implementation meeting are the most detailed public description so far (Dzair Tube). The right-hand column is our reading for industrial companies, not part of the official text.
| Priority | What the source says | What it could mean for industry |
|---|---|---|
| Sovereign AI models | "Sovereign, open artificial intelligence models tailored to the operational needs of the Algerian state" | Aimed at the state first. Signals a preference for open models that can run in the country |
| Computing capacity | Expand national computing so research institutions and start-ups can "develop and train artificial intelligence models domestically" | Possible future access to local GPU capacity; check access terms when they are published |
| Skills | Intensive training with universities and research institutions, especially in data analytics and generative AI | A larger hiring pool, but graduates still need plant knowledge |
| Research to industry | A National AI Research Centre and specialised units to "facilitate the transfer of scientific research into industrial applications" | A route for companies to bring real problems to researchers |
| Data governance | Process and store administrative and public-sector data "within protected national data centres"; ethical, trustworthy AI | Concerns public data, but sets the direction for sensitive data in general |
Some of this is already visible. In March 2025, the ministers of post and telecommunications and of the knowledge economy laid the foundation stone of a high-performance computing centre for AI in Oran, to be equipped with GPUs and to offer computing capacity to researchers, start-ups and Algerian companies (Algérie Eco). On skills, the National Higher School of Artificial Intelligence (ENSIA), created in 2021, graduated 105 students in June 2026 (Radio Algérie).
Which existing laws already apply to industrial AI?
There is no AI-specific law that we know of. Two existing texts matter most for industrial projects. Confirm how they apply to your case with counsel.
Personal data. Law 18-07 of 10 June 2018, amended by Law 25-11 of 24 July 2025, applies to personal data in plant records: names in shift logs, maintenance reports and HR files. According to the CMS legal guide, a controller may transfer personal data to a foreign state only with the authorisation of the national authority (ANPDP). The amendment adds a data protection officer, impact assessments for high-risk processing and breach notification within five days. The same guide reports a 2017 ARPT decision requiring cloud hosting providers to keep customer data on national territory. Our guide to industrial data governance covers this in more detail.
Public procurement. Law 23-12 of 5 August 2023 already favours local capacity (Law 23-12). Article 59 requires a national call for tenders when national production can meet the need. Article 60 requires the specifications, as applicable, to provide for training and the transfer of technology and know-how, and to allow imported products only when no equivalent local product is available or compliant. Article 62 grants a preference margin to products of Algerian origin and to companies under Algerian law majority-owned by resident nationals. Under Article 13, public economic enterprises are not bound by the law's award procedures but must adopt their own, based on its principles. Our AI project RFP guide shows how to use these clauses.
The strategy points in the same direction as these laws: data kept in the country and skills built locally. That is the argument we make in why sovereign AI matters for Algerian industry.
How should industrial companies respond now?
You do not need the final text to act. These steps hold whatever the action plans say.
- Pick use cases on your own data. Start where the payoff and the data are clear: maintenance history, procedures, lab results. An AI readiness assessment helps rank them.
- Classify your data. Decide which documents and datasets may leave the site and which may not, and write it down. Personal data in plant records already falls under Law 18-07.
- Choose architectures that can run in the country. On-premise deployment or local hosting with open-weight models keeps you aligned with the strategy's direction and with the transfer rules. See sovereign cloud, private cloud or on-premise.
- Build skills before you buy. Train operators and engineers by role, and put training and know-how transfer in every AI contract. Our guide to training engineers in AI gives a step-by-step plan.
- Adopt a governance framework now. Until national AI rules exist, use an established one. The NIST AI Risk Management Framework, released in January 2023, is intended for voluntary use (NIST). Its core has four functions: govern, map, measure and manage (NIST AI 100-1).
- Talk to researchers. The research-to-industry priority favours companies that bring real problems and data. ENSIA and university laboratories are natural partners.
- Watch the official channels. Council of Ministers releases, the Ministry of Higher Education and Scientific Research and the Journal officiel will carry the adoption, the action plans and any new rules. Check funding options too; see funding and incentives for digital projects in Algeria.
What we have seen at KABAS AI
We cannot tell you what the action plans will contain. What we can describe is how our own products and published projects line up with the strategy's direction.
On data, the first question to ask of any AI system is where the data goes. DALILE, our multi-agent assistant for technical documents and databases, is built for fully on-premise deployment, so documents and data stay on the client's servers. On skills, our two published projects use Virtual Lab in formulation R&D. At ENAD Shymeca, the implementation began with a discovery phase alongside the client's R&D team, migrated its historical formulation data and included a training and adoption phase for R&D staff. At ADFC, a structured repository of formulation data supports knowledge transfer across generations of technical staff.
Our practical advice: treat the strategy as a direction, not a deadline. Companies that already classify their data, keep sensitive material on site and train their own people will be ready whatever the final text says. The wider context of these changes is in our overview of Algeria's Industry 4.0 transformation, and our comparison of local and international AI vendors helps when choosing a partner.
Where should you start?
Start with one use case and a written data classification. Our consulting team runs AI strategy and readiness assessments covering data, infrastructure, team capabilities and governance. Read more about KABAS AI, or contact us to discuss your sites.
Sources
- Radio Algérie, Conférence africaine des startups : présentation de la première Stratégie nationale de l'intelligence artificielle, 7 December 2024
- Radio Algérie (APS), 3e Conférence africaine des start-up : l'Algérie adopte une stratégie nationale de l'intelligence artificielle, 8 December 2024
- Radio Algérie, Government Meeting: Review of the Draft National Artificial Intelligence Strategy, 25 May 2026
- APS, Réunion de coordination pour la mise en œuvre de la stratégie nationale en matière de l'IA, 5 August 2026
- Dzair Tube, Kamel Baddari launches implementation of Algeria's national artificial intelligence strategy, 5 August 2026
- IT Mag, Kamel Baddari lance la stratégie nationale pour l'intelligence artificielle, 5 August 2026
- Algérie Eco, Algérie : un premier centre de calcul pour l'intelligence artificielle doté de GPU, 17 March 2025
- Radio Algérie, Baddari préside la cérémonie de sortie de deux promotions de l'ENSIA et de l'ENSM, 29 June 2026
- CMS, Expert Guide to Data Protection and Cyber Security Laws: Algeria
- Law 23-12 of 5 August 2023 on public procurement, Journal officiel no. 51 (FAOLEX copy)
- NIST, AI Risk Management Framework
- NIST, Artificial Intelligence Risk Management Framework (AI RMF 1.0), NIST AI 100-1, January 2023
Frequently asked questions
Has Algeria adopted a national AI strategy?
Partly. A first version was presented in December 2024. In May 2026 the government reviewed a draft strategy that still had to be adopted by the Council of Ministers, and on 5 August 2026 a first meeting launched its implementation. At the time of writing we have not found a published full text, so treat details not stated in official releases as provisional.
Is there an AI law in Algeria?
We are not aware of an AI-specific law. Existing texts already apply to AI projects, notably Law 18-07 on personal data, amended by Law 25-11 of 24 July 2025, and Law 23-12 on public procurement. Confirm how they apply to your project with legal counsel.
Does the strategy require companies to host AI in Algeria?
Not as published so far. The August 2026 priorities speak of storing administrative and public-sector data in protected national data centres, which concerns the public sector. For companies, the rule that already matters is that transfers of personal data abroad need the authorisation of the national data protection authority.
Which ministry leads Algeria's AI strategy?
The first version was drafted by the Scientific Council for Artificial Intelligence, an advisory body supervised by the ministries of higher education and of the knowledge economy and start-ups. The government meeting of May 2026 was chaired by the Prime Minister, and the first implementation meeting, in August 2026, by the Minister of Higher Education and Scientific Research.
Can industrial companies use national AI computing capacity?
Possibly in the future. The high-performance computing centre for AI whose foundation stone was laid in Oran in March 2025 is described as serving researchers, start-ups and Algerian companies. We have not found published access terms or prices, so ask before planning a project around it.



