Why Ghanaian Businesses Are Slowly Adopting AI — The 85th Global Ranking, the GH¢500bn Target, and the Credibility Gap That Could Undermine It All

Why Ghanaian Businesses Are Slowly Adopting AI — Ghana ranks 85th globally in AI readiness. Less than 5% of SMEs use AI. The GH¢500bn GDP target is ambitious, but infrastructure gaps, data costs, skills shortages, the Publican AI crisis, and policy‑execution gaps are slowing adoption. Our deep‑dive analysis reveals six barriers and three scenarios for Ghana’s AI future.

Executive Introduction

The ambition is unmistakable. On 24 April 2026, President John Dramani Mahama formally launched Ghana’s National Artificial Intelligence Strategy (2025–2035), projecting that AI would contribute GH¢500 billion (approximately 250 million commitment for a national AI computing centre and an additional $20 million for short-to-medium-term implementation.

Yet the distance between that ambition and the current reality on the ground is sobering. In the 2025 Oxford Insights Government AI Readiness Index, Ghana ranked 85th out of 188 countries globally, behind Kenya (65th), South Africa (67th), Mauritius (71st), Nigeria (72nd), and Rwanda (75th). In the Global AI Index measuring investment, innovation and implementation, Ghana placed 61st, with very low scores across the development, research, business ecosystem, talent and infrastructure sub-pillars. The African AI Governance Index placed Ghana in the “emerging” category with a score of 35.7, below several continental peers.

This profile examines why Ghanaian businesses are slowly adopting AI across six interconnected barriers: the infrastructure gap that turns ambition into a compute desert; the data cost barrier that prices entire segments of the workforce out of the AI economy; the skills deficit masked by the One Million Coders programme; the trust deficit that erupted violently at Ghana’s ports; the SME adoption chasm; and the policy-execution gap that threatens the entire strategy. The strategy is launched. The computing centre is coming. The institutions are being built. But for now, the distance between ambition and execution remains the single most important fact about Ghana’s AI economy.

The Infrastructure Gap — Ambition Without Compute

Artificial intelligence demands a fundamentally different kind of digital infrastructure from the connectivity that powered Ghana’s previous digital transformation. For more than a decade, Ghana’s digital progress has been defined by connectivity: broadband networks expanded, mobile penetration increased, and digital financial services scaled rapidly. But AI is forcing the next phase, defined not by access, but by where computation happens. Countries that cannot host their own compute will increasingly export not just data, but value, control and strategic capability.

The gap is stark. At the launch of the National AI Strategy, the government demonstrated a chatbot named “Aku AI” intended to showcase progress. However, the demonstration highlighted the gap between government systems and what the private sector has already achieved, raising concerns about the system’s ability to respond in real time despite being presented as a functional AI tool capable of interacting in local languages. This was not a failure of technical expertise but rather institutional delays in adopting and integrating emerging technologies. The private sector and academic institutions in Ghana are already deploying advanced AI solutions across various sectors — including Farmerline, mPharma, and KNUST’s medical imaging work — demonstrating that the issue is not capability but the pace at which public institutions are adapting to innovation.

The infrastructure constraint is increasingly clear. Power reliability remains uneven. Interconnection ecosystems are still maturing. Enterprise workloads are often routed externally, even when local capacity exists. Policy ambition, while rising, is not yet fully matched by execution frameworks. The government’s response — the $250 million AI computing centre — is ambitious, but its operational timeline remains uncertain, and experts have warned that without clear procurement principles, the centre risks becoming a symbolic achievement rather than a functional asset.

The result is a widening gap between demand and capacity. Demand is already emerging from financial institutions, telecom operators and public‑sector platforms integrating AI into core operations. At the same time, regulatory signals are shifting, with the Bank of Ghana’s push toward localisation of critical financial workloads reflecting a broader trend: governments are no longer treating digital infrastructure as a neutral utility, but as a sovereign asset. Yet infrastructure is not scaling at the same pace.

The Data Cost Barrier — The Foundation That Is Not Yet Affordable

Before businesses can adopt AI, they need reliable, affordable connectivity. By that measure, Ghana is not ready.

Former Vice President Dr Mahamudu Bawumia has warned that the cost of mobile data remains a significant barrier to Africa’s participation in the AI revolution. Focusing on Ghana, he indicated that the cost of 1GB of data ranges between approximately $0.05 and $1.50, depending on the provider and bundle. While this positions Ghana relatively well compared to peers, the disparity in real access is stark: internet access is relatively affordable for middle- and high-income groups, however, it remains costly for low‑income households, largely due to income inequality and the structure of the informal sector.

The implication for AI adoption is direct. Training AI models requires massive amounts of data. Using AI tools requires constant connectivity. If connectivity remains too expensive for the majority of Ghanaian businesses and households, AI adoption will be concentrated among large corporations and affluent urban consumers. As Bawumia stated: “Before we debate algorithms, we must be disciplined about the foundations that enable adoption at scale. Without affordable connectivity, AI risks benefiting only a narrow segment of the population”.

The government has responded by mandating that telecom operators increase data bundle values without raising prices — a measure that treats the symptom, not the cause, of unaffordable connectivity. The 39 per cent tax burden on telecommunications services remains largely untouched. The dedicated electricity tariff for telecom towers has not been implemented. The “Dig Once” policy for fibre rollout remains a proposal, not a reality. Until these structural barriers are addressed, data will remain too expensive for the businesses that need AI most — the small and medium enterprises that constitute the majority of Ghana’s economy.

The Skills Deficit — The Coders Who Cannot Find Jobs

Ghana’s flagship skills initiative, the One Million Coders Programme, aims to train one million Ghanaians in digital and AI skills at scale, with 300,000 targeted for training in 2026 alone. The programme is designed to reskill, upskill, and skill Ghanaians from being digitally native to becoming AI native. Fifty digital hubs are being established nationwide to support the training. On paper, this is a world‑class response to the skills shortage.

But critics warn that the numbers mask a deeper problem. Policy analyst and founding president of IMANI Africa, Franklin Cudjoe, has warned that Ghana risks falling behind in the global AI revolution due to fragmented and politically driven skills programmes that lack clear strategic direction. “We are in an AI revolution, and we are doing what? One million coders. I don’t even know what that means. These programmes are crowded, they are all over the place”. Cudjoe argued that many of the skills promoted under programmes such as coding can easily be learned online and should not be treated as standalone flagship policies unless they are closely linked to industry demand and employment opportunities.

The scale of the employment challenge puts the skills deficit in perspective. In a recent security services recruitment exercise, more than 405,000 applicants applied for just 5,000 available positions — figures that have reignited debate about youth unemployment and job creation in Ghana. The country’s security services collectively employ fewer than 100,000 personnel, with their combined wage bill already nearing GH¢13 billion — highlighting the fiscal limits of relying on public sector recruitment to absorb large numbers of job seekers.

Dr Bryan Acheampong, Member of Parliament for Abetifi, has called for a comprehensive National Skills Plan for the AI Transition, warning that without it, Ghana risks widening inequality and missing out on the transformative potential of AI in job creation. He proposed that such a plan be developed through coordinated efforts involving the Ministry of Education, Youth Employment Agency and private sector partners, including vocational training, digital apprenticeships, scholarships, micro‑credentials, and community‑based AI awareness programmes. This is not just about STEM, he argued, but about teaching students how to use AI in the fields they live and work in.

In the absence of a coordinated skills plan, Ghanaian businesses face a chronic shortage of AI‑capable talent. The talent that exists is concentrated in Accra and a few other urban centres. The businesses that need AI most — the agricultural SME in the Northern Region, the logistics company in Kumasi, the healthtech startup in Takoradi — cannot find the talent they need at prices they can afford.

The Trust Deficit — The Publican AI Crisis at the Ports

The most dramatic evidence of the trust deficit in Ghana’s AI adoption unfolded at the country’s ports in early 2026. The Ghana Revenue Authority (GRA) rolled out the Publican AI system, an AI‑driven customs valuation and tariff classification system designed to monitor valuations and detect misclassification of goods. The government justified the move by claiming it was losing significant revenue due to improper declarations by importers.

The backlash was immediate and severe. The Ashanti Business Owners Association (ABOA) called for the immediate suspension of the system, describing it as operating largely as a “black box,” making it difficult for importers to understand how values and tariff classifications are determined. This, they argued, undermines due process and threatens fair trade practices. Without clear justification or a standardised appeal mechanism, businesses are left vulnerable to unnecessary financial burdens.

Spare parts dealers in Kumasi mounted strong resistance, describing the system as a major threat to their survival. The Suame Magazine Chairman stated that the AI system has significantly increased import duties, making it difficult for dealers to stay in business, with some already on the verge of shutting down. The dealers warned that failure by government to abolish the system will compel them to increase spare parts prices by 50 to 70 per cent in order to survive.

The former President of the Ghana Union of Traders’ Associations (GUTA) added his voice, warning that the system could increase the cost of doing business and disrupt trading activities, noting that some importers are already experiencing higher duties. The GUTA Director of Welfare revealed that major stakeholders were left out of the consultation process prior to the system’s implementation, with the union only becoming aware of the initiative when the Finance Minister mentioned it in the 2026 budget. In response to the growing opposition, ABOA proposed a hybrid model where AI serves only as a support tool subject to validation by qualified customs officers, with clear guidelines on how AI‑generated values are derived and a transparent dispute resolution system.

The Publican AI crisis is not an isolated case of regulatory overreach. It is a warning. If Ghana’s government deploys AI without transparency, without consultation, and without grievance mechanisms, businesses will not trust AI systems. And without trust, adoption will stall before it starts. Imani Africa’s analysis of the crisis concluded that “judgment rather than replace it, Ghana can bridge its revenue gap without sacrificing the survival of its trading community”. The lesson is clear: AI systems must augment human decision‑making, not replace it opaquely. Importers and customs officers must be able to understand, challenge and appeal AI‑generated valuations. Without these safeguards, AI becomes a tool of extraction, not a tool of efficiency.

The SME Adoption Chasm — Less Than 5 per cent Using AI

For all the talk of national AI strategies and computing centres, the vast majority of Ghanaian businesses are not using artificial intelligence. They are not even close.

A peer‑reviewed study of technology adoption among Ghanaian SMEs found that close to 60 per cent of surveyed businesses had adopted less sophisticated technologies such as social media, mobile banking and e‑commerce. But overall, less than 5 per cent of the SMEs surveyed were using sophisticated technologies such as AI and the Internet of Things (IoTs). The study identified the top five barriers to technology adoption: lack of finance, insufficient government support, lack of absorptive capability, insufficient demand, and poor technology enabling infrastructure.

Another survey found that while over 92 per cent of SMEs use mobile payment systems and mobile money — which are now essential for business operations — cloud storage and collaboration tools lag far behind at just 1.6 per cent adoption. The pattern is consistent: Ghanaian SMEs adopt technologies that are essential for survival — mobile payments, social media marketing — but they do not adopt technologies that require significant investment, technical expertise, or organisational change.

The SME adoption gap has direct implications for Ghana’s AI ambitions. The National AI Strategy projects that AI will contribute GH¢500 billion to GDP by 2035. But if less than 5 per cent of SMEs — the backbone of Ghana’s economy, accounting for more than 90 per cent of all businesses — are using AI, that target will remain a fantasy. The SMEs that constitute the majority of Ghana’s economy are not being reached by the AI revolution, and the policies designed to reach them are not yet operational.

This gap between consumer and business adoption is visible across the digital economy. While 95 per cent of individuals have used digital payments as consumers, only 37 per cent of businesses accept or use digital payment platforms. Adoption remains heavily concentrated in Greater Accra and a few regional capitals, with knowledge gaps, fraud fears and uncertainty over returns limiting broader adoption, especially among smaller enterprises and in the agricultural sector.

The Policy‑Execution Gap — Strategy Without Implementation Architecture

The National AI Strategy is a document of genuine ambition. It targets a GH¢500 billion contribution of AI to GDP by 2035 and is built around eight pillars covering education, infrastructure, data governance, innovation, and public sector adoption. But experts have identified critical gaps that could undermine its implementation.

Desmond Israel, founder and lead consultant at Information Security Architects Ltd, identified what he described as a mechanical revision of an earlier policy framework, pointing to sections of the strategy that retain language referencing a 2033 endpoint, consistent with a previous 2023 to 2033 draft, despite the document being positioned as a fresh ten‑year roadmap through 2035. “National strategies are built around timelines. If those timelines are inconsistent within the same document, it weakens confidence in execution planning,” he said.

On governance, Israel raised concerns about the strategy’s institutional architecture. The document references a Responsible AI Authority, a Responsible AI Office, and a National AI Office without defining their respective mandates or how they would relate to one another. “AI policy is governance infrastructure, not branding. Without clearly defined institutions, implementation risks becoming fragmented”.

Hector Dotse, a cybersecurity analyst, identified three major decisions that remain deferred. The strategy sets a GH¢200 billion private investment target but does not outline the instruments required to mobilise that capital. It also stops short of specifying a regulatory model, leaving open whether Ghana will adopt a risk‑based framework similar to the EU, a sector‑led approach like the UK, a state‑coordinated model comparable to China, or sandbox‑driven regulation as practised in Singapore. Additionally, Dotse noted the absence of procurement principles ahead of major infrastructure commitments, including the planned national AI compute centre.

Abdul‑Mumin Sofo Yumzaa, Executive Director of Simba Ghana, raised concerns over Ghana’s preparedness to effectively implement the strategy. He observed that private sector actors and academic institutions in Ghana were already deploying advanced AI solutions — citing Farmerline, mPharma and KNUST — and concluded that the issue was not capability but the pace at which public institutions are adapting to innovation. He contrasted Ghana’s progress with countries such as Rwanda and Senegal, which adopted national AI frameworks earlier and had since integrated the technology into their broader development agenda.

Yumzaa also cautioned against policy and regulatory risks, pointing to the proposed Emerging Technologies Bill, which sought to regulate AI, robotics, and blockchain, warning that excessive controls could stifle innovation. He expressed concern about possible data localisation requirements, which could increase operational costs for startups and limit their competitiveness beyond Ghana. Despite these concerns, he noted that the planned adoption of AI across more than 100 Ministries, Departments and Agencies from 2026 could create a substantial market for local technology firms.

The central challenge, as Dotse concluded, is that “none of these requires reopening the strategy. All of them require follow‑on documents in the coming months.” The principal risk is not outright failure but gradual misalignment between the strategy’s ambitions and its implementation architecture.

The Future Outlook — Three Scenarios for AI Adoption in Ghana

Scenario One — Gradual, Concentrated Adoption (65 per cent probability).

AI adoption remains concentrated in large corporations, financial institutions, telecom operators and foreign‑owned enterprises. SMEs continue to lag, with adoption remaining in single digits. The computing centre becomes operational but serves primarily research institutions and large enterprises. The One Million Coders Programme produces entry‑level graduates but lacks the advanced training needed for high‑value AI roles. The Publican AI crisis is resolved through a hybrid model, but trust in government‑led AI remains fragile. AI contributes GH¢100‑150 billion to GDP by 2035 — significant, but far below the GH¢500 billion target.

Scenario Two — Accelerated Breakthrough with Sustained Investment (25 per cent probability).

The computing centre becomes operational by 2028 and attracts foreign investment. The National Skills Plan for the AI Transition is implemented effectively, linking training to industry demand. SMEs begin adopting AI through cloud‑based solutions and low‑cost platforms. The Publican AI crisis leads to a transparent, accountable regulatory framework that rebuilds trust. Open data policies unlock government datasets for private sector innovation. AI contributes GH¢300‑400 billion to GDP by 2035 — approaching the target.

Scenario Three — Implementation Stagnation (10 per cent probability).

The computing centre faces technical and financial challenges, with its operational timeline slipping. The One Million Coders Programme produces graduates who cannot find AI jobs. Skills shortages persist, and talent migrates abroad. The Publican AI crisis is not resolved, and trust in government AI systems collapses. Infrastructure investments are delayed, and data costs remain high. AI adoption stalls, and the gap between Ghana and regional peers widens.

The most likely path is Scenario One: gradual, concentrated adoption, with AI transforming the largest enterprises but leaving SMEs and the informal sector largely untouched. The GH¢500 billion target is not impossible, but it is improbable without sustained, coordinated action across infrastructure, skills, trust and policy execution. The strategy is launched. The institutions are being built. The funding is committed. But the distance between ambition and execution remains the single most important fact about Ghana’s AI future.

Conclusion

Why are Ghanaian businesses slowly adopting AI? The answer is not a single barrier but a matrix of them. Infrastructure is insufficient: power is unreliable, data centre capacity is limited, and the $250 million computing centre is not yet operational. Data costs remain too high, pricing low‑income households and small businesses out of the AI economy. Skills are concentrated and shallow: the One Million Coders programme produces graduates who cannot find AI jobs, and there is no coordinated skills plan linking training to industry demand. Trust is fragile: the Publican AI crisis demonstrated that when AI is deployed without transparency, consultation and appeal mechanisms, businesses will resist, not adopt. SMEs are structurally excluded: less than 5 per cent are using AI, and the barriers — lack of finance, insufficient government support, poor infrastructure — are not being addressed. And the policy‑execution gap remains wide: the National AI Strategy is ambitious, but its timelines are inconsistent, its institutional architecture is undefined, and its critical decisions on regulation, procurement and capital mobilisation are deferred.

The government has taken necessary first steps. The National AI Strategy is launched. The computing centre is funded. The One Million Coders programme is training digital skills. The Bank of Ghana’s CISD 2026 has established AI governance standards for the financial sector. But first steps are not final steps. The distance between the GH¢500 billion target and the current reality — 85th in global AI readiness, less than 5 per cent of SMEs using AI, a labour force adding 500,000 entrants annually — is vast. Closing that gap requires not more strategy documents but implementation architecture: clear procurement principles for the computing centre, a coordinated skills plan linking training to jobs, transparent AI deployment frameworks that rebuild trust, and affordable connectivity that reaches every business, not just the largest corporations.

The algorithm is not coming to Ghana’s economy. It is already here, in the ports, in the banks, in the chatbots. The question is not whether AI will be adopted. It will be. The question is whether Ghana’s businesses — especially the SMEs that constitute the backbone of the economy — will be ready when it arrives. The country’s AI future will be determined not by the ambition of its strategy but by the credibility of its execution. The strategy is launched. Now comes the hard part. The businesses of Ghana are watching. And most of them, for now, are still waiting to see if the AI revolution will include them — or leave them behind.

Frequently Asked Questions (FAQ)

Q1: How many Ghanaian businesses are currently using AI?

Less than 5 per cent of small and medium enterprises surveyed in Ghana are using sophisticated technologies such as AI and the Internet of Things (IoTs). Most businesses — approximately 60 per cent — have adopted less sophisticated technologies such as social media, mobile banking and e‑commerce, but adoption of advanced AI remains negligible outside large corporations.

Q2: What is Ghana’s National AI Strategy and when was it launched?

Ghana’s National Artificial Intelligence Strategy (2025‑2035) was launched on 24 April 2026 by President John Dramani Mahama. It projects that AI will contribute GH¢500 billion (approximately $45 billion) to Ghana’s GDP by 2035 and is built around eight pillars covering education, youth employment, digital infrastructure, data governance, ecosystem development, sectoral AI adoption, applied research, and ethical AI governance.

Q3: How much is Ghana investing in AI infrastructure?

The government has committed $250 million to establish a national AI computing centre and an additional $20 million for short‑to‑medium‑term implementation of the National AI Strategy. The computing centre is intended to serve as a hub for research, innovation and enterprise development, enabling local talent to create solutions with relevance beyond Ghana’s borders.

Q4: What is the Publican AI system and why did it cause a crisis?

The Publican AI system is an AI‑driven customs valuation and tariff classification system deployed by the Ghana Revenue Authority at the country’s ports in early 2026. Importers and business owners objected to its lack of transparency — describing it as a “black box” — its unpredictable valuations, the absence of a clear appeal mechanism, and the fact that major stakeholders were not consulted before its rollout. Some dealers threatened to increase prices by 50‑70 per cent in response to higher import duties.

Q5: How does Ghana’s AI readiness compare to other African countries?

Ghana ranks 85th out of 188 countries globally in the 2025 Oxford Insights Government AI Readiness Index, behind Kenya (65th), South Africa (67th), Mauritius (71st), Nigeria (72nd), and Rwanda (75th). In the Global AI Index measuring investment, innovation and implementation, Ghana ranks 61st, with very low scores in development, research, talent, infrastructure and business ecosystem.

Q6: What is the One Million Coders Programme?

The One Million Coders Programme is Ghana’s flagship skills initiative, aimed at training one million Ghanaians in practical digital and AI skills at scale. In 2026 alone, 300,000 young Ghanaians are expected to be trained. Fifty digital hubs are being established nationwide. Critics, however, argue that the programme lacks clear strategic direction and is not closely linked to industry demand and employment opportunities.

Q7: How does the cost of data affect AI adoption in Ghana?

The cost of 1GB of mobile data ranges from approximately $0.05 to $1.50, depending on the provider and bundle. While this compares favourably to many peers, affordability remains a challenge for low‑income households, largely due to income inequality and the structure of the informal sector. High data costs limit the ability of small businesses and low‑income individuals to access AI tools and services.

Q8: What barriers prevent SMEs from adopting AI in Ghana?

A peer‑reviewed study identified the top five barriers: lack of finance, insufficient government support, lack of absorptive capability (the ability to learn and implement new technologies), insufficient demand, and poor technology enabling infrastructure. Less than 5 per cent of surveyed SMEs were using AI or IoTs.

Q9: What is the Bank of Ghana’s role in AI governance?

The Bank of Ghana introduced a revised Cyber and Information Security Directive (CISD 2026) in March 2026, which includes new governance standards for artificial intelligence and machine learning systems used in fraud detection and credit scoring, aimed at ensuring transparency and security in automated decision‑making. The directive extends regulatory coverage to fintechs, microfinance institutions and other financial sector players.

Q10: What is the outlook for AI adoption in Ghana’s businesses?

The most likely scenario is gradual, concentrated adoption. AI will transform large corporations and financial institutions, but SMEs — which constitute more than 90 per cent of registered businesses — will remain largely untouched. Infrastructure gaps, data costs, skills shortages, trust deficits and policy‑execution gaps will slow progress. The GH¢500 billion GDP target is achievable only with sustained, coordinated action across all six barriers.

Q11: How does Ghana’s National AI Strategy address trust and transparency?

The strategy places digital sovereignty and ethical governance at its core, prioritising citizen trust through ethical AI adoption, data protection and robust regulatory frameworks. However, experts have noted that the strategy does not specify procurement principles for the computing centre, does not define the mandates of its proposed governance institutions, and has not yet established a clear regulatory model for AI deployment.

Q12: What should Ghanaian businesses do to prepare for AI?

Businesses should focus on digitising core operations first — moving from paper to digital records, adopting cloud‑based tools, and building data infrastructure. They should identify specific business problems that AI could solve — inventory optimisation, customer service automation, credit scoring — rather than pursuing AI as a general‑purpose tool. And they should invest in basic data literacy and AI awareness for their staff, recognising that AI adoption is an organisational change, not just a technology purchase.

Esther Aku-Sika

Esther Aku-Sika is a content writer and social media strategist who helps brands and startups grow through intentional storytelling and practical marketing strategies. With a keen eye for trends and audience behavior, she shares business insights, content strategies, and real-life lessons to help entrepreneurs build visibility and turn ideas into income. Through her writing, she simplifies complex concepts and equips readers with actionable steps to grow in today’s digital space.