Part 1: Getting the Foundations Right
“The foundations of AI are not algorithms; they are infrastructure, data, and people.”
Ghana deserves praise for launching its National Artificial Intelligence (AI) Strategy. For the first time, the country articulated a comprehensive vision to harness AI to transform the economy, improve public services, create opportunities for young people, and position Ghana as a leader in Africa’s digital future. The strategy is ambitious. It seeks to make Ghana an AI-powered society, strengthen digital infrastructure, develop AI talent, support innovation, improve public sector efficiency, and unlock new sources of economic growth.
The strategy is structured around eight pillars, including AI education and training, youth empowerment, digital infrastructure, data access and governance, AI ecosystem development, sector adoption, applied research, and public sector transformation.
Undoubtedly, Ghana’s AI strategy captures the right aspirations reflecting the hard work of policymakers, researchers, industry leaders, and stakeholders who invested time in developing a national roadmap for AI, and they deserve commendation.
However, Artificial intelligence will not transform Ghana because a strategy document exists. AI will transform Ghana when we build the requisite foundations and move beyond strategy to execution. In a three part series, I aim to draw policy makers and the public attention to the precursors to implementing a successful AI strategy.
In this first part, I provide a background to Ghana’s AI journey, a description of the foundations of a successful AI strategy, and a call to action beyond strategy documents.
Ghana’s AI Journey Did Not Begin in 2026
As we celebrate the National AI Strategy, it is important to recognize that it is not the beginning of Ghana’s AI journey. Before AI became a mainstream policy discussion, a growing ecosystem of students, researchers, entrepreneurs, engineers, and practitioners had already begun laying the foundations for Ghana’s AI future, and I am privileged to be part of that journey
In 2019, Augustine Denteh (PhD) and I launched the maiden Ghana Data Science Summit. At the time, conversations about machine learning, artificial intelligence, and data science were nonexistent in Ghana. Our objective was simple but ambitious: create awareness, build local capacity, and expose students and professionals to the promise of data science for economic transformation.
The summit brought together academia, industry, government, and the technology community to discuss data management, analytics, machine learning, and artificial intelligence. More importantly, it fostered a community around these technologies.
In 2021, when COVID-19 disrupted the world, we organized the summit virtually and continued providing learning opportunities, mentorship, and community engagement for aspiring data scientists and AI practitioners.In 2022, as global excitement around AI accelerated, we deliberately shifted discussions beyond hype and focused on practical adoption. We explored the realities of data access, digital infrastructure, governance, responsible AI, and the need to build solutions tailored to Ghanaian problems rather than simply importing technologies from elsewhere.The year 2023 marked a historic milestone when Ghana hosted the Deep Learning Indaba, Africa’s premier artificial intelligence gathering.
Researchers, innovators, policymakers, entrepreneurs, and students from across Africa and beyond came together for technical workshops, lectures, research presentations, mentorship programs, and policy discussions., Ghana became the center of AI conversations on the continent.In 2024, the Ghana Data Science Summit moved to KNUST in Kumasi in partnership with the Responsible AI Lab. The goal was clear: AI development in Ghana must not be concentrated in Accra. It must become a national movement that reaches students and innovators across the country. In 2025, the summit continued at Ashesi University, focusing on practical AI, technical capacity building, entrepreneurship, and ecosystem development.
This year, the summit is taking place in Ho from June 24- 26 under the theme “Data First: Unlocking Ethical and Inclusive AI in Ghana.” The theme is deliberate. It reflects a reality that is often overlooked in AI discussions: before AI comes data. Without data, there is no AI.The 2026 program reflects that practical orientation.
It includes mentorship sessions on research, graduate school, data careers, entrepreneurship in AI and data, and building a strong professional portfolio. It includes hands-on tutorials in Python, SQL, data engineering, machine learning, synthetic data generation, ethical large language model assistants, agentic AI systems, interpretable machine learning, ethical vision systems, and machine learning in production. It also includes panels on data access, data science jobs in Ghana, and the realities of building AI for Ghanaian contexts.
Across the seven years, the Ghana Data Science Summit, IndabaX Ghana, and related initiatives have impacted more than a thousand students, researchers, and professionals through workshops, tutorials, mentorship, policy discussions, and industry collaborations. This history matters because it teaches us that AI transformation is not achieved by only strategy documents. It is achieved through sustained investment in people, communities, education, infrastructure, research, experimentation, and collaboration.
Ghana Is Not Starting From Zero
Ghana is not starting from scratch. There are already signs of what is possible when local talent builds for local realities.One strong example is GhanaNLP and Khaya AI. Ghanaian researchers and technologists have worked on language technologies for Ghanaian and African languages, including Twi, Fante, Ewe, Ga, Kusaal, Dagbani, and others. GhanaNLP’s parallel corpora project has developed more than 41,000 aligned sentence pairs across English and five Ghanaian languages, helping address the lack of digital resources for local-language AI.
Khaya, one of the visible products from this ecosystem, supports translation, speech recognition, and text-to-speech capabilities for Ghanaian and African languages.This is the kind of AI Ghana needs: AI that understands our languages, our context, and our people. Another example is the growing work around AI for education, including Kwame for Science and Adesua, a WhatsApp-based AI teaching assistant for science learning in West Africa. These tools show how AI can be adapted to platforms that students already use, such as WhatsApp, and applied to real educational challenges, including limited access to personalized learning support.
These examples prove that Ghana’s AI future does not have to be imported wholesale. It can be built here, by Ghanaians and Africans, for Ghanaian and African problems.
The Four Foundations That Will Determine Success
Among the various pillars of Ghana’s AI Strategy, four foundational enablers deserve special attention: Digital Infrastructure, AI Education and Training, Youth Empowerment, and Data Access and Governance. Without these foundations, the other pillars will remain aspirations.
1. Digital Infrastructure: AI Cannot Run on Weak Foundations
AI requires reliable digital infrastructure including electricity, internet connectivity, computing infrastructure, and cloud services. Unfortunately, many realities facing ordinary Ghanaians reveal how much work remains.
National internet access has improved significantly. DataReportal reported that Ghana had about 24.3 million internet users in January 2025, representing 69.9 percent internet penetration, and later reported 26.3 million users at the end of 2025, or 74.6 percent penetration. Those numbers are encouraging.
But national averages can hide deep inequalities. A World Bank analysis found that internet uptake in urban Ghana rose to about 80 percent, while rural uptake rose to about 54 percent. That rural figure matters. If nearly half of rural citizens are still outside effective internet use, then AI risks becoming an urban privilege rather than a national development tool.
Even where connectivity exists, meaningful access remains a challenge. Data can still be expensive relative to income for many households. Internet quality and speed are inconsistent. Many students, entrepreneurs, and small businesses struggle not only to get online, but to stay online long enough and at a quality high enough to learn, build, collaborate, and use modern AI systems.
This distinction is important. Having internet access is not the same as having meaningful digital access. A student in Accra and a student in a rural district may both technically have internet, but their ability to take an online AI course, download learning materials, join video sessions, build products in the cloud, or participate in digital work can be vastly different.
If Ghana wants AI to transform agriculture, healthcare, education, financial services, and public administration, digital infrastructure must become more than an urban advantage. It must become a national public good. That means investing not only in network expansion, but also in affordability, reliability, broadband competition, rural connectivity, local computing infrastructure, and stable electricity.
2. AI Education and Training: We Need Builders, Not Just Users
The strategy rightly emphasizes AI education. But Ghana’s objective should not simply be to create users of AI tools. We must create builders, researchers, engineers, product managers, entrepreneurs, policymakers, and innovators.
AI education must extend beyond prompt engineering and chatbot usage. It should include computational thinking, statistics, machine learning, data engineering, software engineering, cloud computing, product development, ethics, cybersecurity, and responsible AI.
At the basic level, students should develop digital literacy and problem-solving skills. At the tertiary level, universities and technical institutions must modernize curricula to reflect today’s AI realities. And at the professional level, Ghana must create continuous learning pathways for workers whose roles will be transformed by AI.
This cannot be done by theory alone. The country must involve practitioners who have built and deployed real AI systems, researchers who understand science, educators who understand pedagogy, and domain experts who understand local problems. The Ghanaian diaspora should also be intentionally mobilized. There are Ghanaians building advanced AI systems across the world; their expertise should be connected to national capacity building.
If we get AI education right, Ghana will build capability. If we get it wrong, we will simply produce certificates.
3. Youth Empowerment: Beyond the One Million Coders Vision
The ambition to create one million AI-ready youth is commendable. Ghana’s youth are creative, energetic, and eager for opportunity. But training alone is not enough.
The goal should be one million problem-solvers, not simply one million coders.
The world is changing rapidly. AI systems are already writing code, generating software, automating analysis, and accelerating product development. Therefore, a national youth AI program cannot focus only on teaching young people old-style coding. It must teach them how to identify problems, understand users, work with data, build products, evaluate AI systems, create businesses, and deploy solutions responsibly.
The urgency is real. Recent labour statistics reported by Ghana Statistical Service indicate that more than 1.3 million young Ghanaians aged 15 to 24 were not in employment, education, or training in Q3 2025, representing about 21.5 percent of that age group. Youth unemployment and underutilization remain major national challenges.
AI readiness must therefore be connected to jobs, startups, apprenticeships, research opportunities, internships, seed funding, and industry partnerships. We should not train thousands of young people only to leave them unemployed. Training must be connected to opportunity.
A strong youth empowerment program should include technical training, product design, entrepreneurship, AI ethics, mentorship, innovation challenges, startup incubation, and funding for youth-led solutions in agriculture, healthcare, education, logistics, public service, and financial inclusion.
4. Data Access and Governance: The Most Important Enabler
This may ultimately be the most important pillar in the entire strategy.
Artificial intelligence runs on data. Without data, there is no AI. Without quality data, there is no useful AI. Without accessible data, there is no innovation. And without governance, there is no trust.
Most conversations about AI governance focus on regulation, privacy, ethics, and compliance. These are important. But before we govern data, we must ask a more fundamental question: do we actually have the data required to build AI systems that solve Ghana’s problems?
Many institutions still operate on paper-based processes. Data is often fragmented across agencies, stored in incompatible systems, trapped in PDFs, locked in filing cabinets, or difficult to access. Researchers, startups, and innovators frequently struggle to obtain even non-sensitive datasets that could be used to create value.
Before governance comes access. Before machine learning comes data collection. Before predictive analytics comes digitization.
We cannot talk seriously about AI in healthcare when patient information is fragmented. We cannot talk about AI in education when anonymized educational data is difficult for researchers to access. We cannot talk about AI in agriculture when high-quality production, pricing, climate, logistics, and market data are not easily available to innovators. We cannot talk about AI-powered government when key public-sector processes remain analogue.
This does not mean abandoning governance. On the contrary, strong governance is essential. Privacy, cybersecurity, data standards, interoperability, access protocols, anonymization, and accountability must all be strengthened. But governance should enable responsible innovation rather than unintentionally limiting it.
Ghana must treat data as national infrastructure. We need digitization, standardized data collection, interoperable systems, responsible data sharing mechanisms, secure platforms, and clear rules that allow appropriate access while protecting citizens.
Moving Beyond Documents to Execution
The National AI Strategy provides direction. It provides ambition. It provides a framework. But strategies do not create transformation. Execution does.
Every major AI initiative should have clear objectives, measurable outcomes, transparent reporting, and accountable leadership. Citizens should know what problem is being solved, what investment is being made, who is responsible, what outcomes are expected, and how success will be measured.
The government should work with real experts. This includes academia, industry practitioners, startups, civil society, domain experts, students, and the Ghanaian diaspora. Contracts and programs should be awarded based on competence and measurable outcomes, not connections.
We must also adopt a product-building mindset. End users should be involved from the beginning. Farmers, nurses, teachers, traders, students, civil servants, small business owners, and community leaders should help shape the AI systems meant to serve them.
If we invest in AI without measuring impact, we will repeat the old pattern: big announcements, limited results. We need public dashboards, annual progress reports, independent audits, and clear accountability for each pillar of the strategy.
Conclusion
Ghana’s National AI Strategy is an important milestone. Those who contributed to it deserve recognition. It demonstrates that Ghana understands the importance of artificial intelligence in shaping the future.
Yet an AI-powered Ghana cannot be built on expensive data, intermittent connectivity, fragmented datasets, and analogue processes. The foundations of AI are not algorithms; they are infrastructure, data, and people.
If Ghana gets these foundations right, the country can become one of Africa’s true AI success stories. The talent exists. The vision exists. The momentum exists.
What matters now is disciplined execution.
This is Part 1 of this series. In Part 2, I will examine the accelerator pillars of the strategy, including AI ecosystem development, sector adoption, applied research, and public sector transformation, and discuss how Ghana can convert ambition into measurable national impact.
Readers should look out for Part 2.
–
Author: Delali Agbenyegah, Co-Founder, Ghana Data Science Summit | Lead Organizer, IndabaX Ghana | Senior Data Science & Engineering Lead, Shopify USA | Advisor, Wallbridge AI
