But a new report by Column reveals that despite this surge of activity, there are weaknesses that affect much of Africa’s AI policy space: great vision, but little follow-through.

According to the State of AI Policy in Africa 2025 report, authored by Mo Shehu and Gideon Onunwa, Nigeria scores 18 out of 24 on the AI Governance Maturity Index, placing it in the continent’s second tier. 

The country has innovation and global attention, but lacks a dedicated budget, enforceable law, or monitoring framework.

“The findings show a clear divide between ambition and execution,” the authors write.

Promise Without Policy Backbone

Nigeria’s draft National Artificial Intelligence Strategy (2024) rests on five pillars: infrastructure, ecosystem, adoption, responsible AI, and governance. Oversight sits with the Federal Ministry of Communications, Innovation and Digital Economy and the National Centre for Artificial Intelligence & Robotics (NCAIR).

The document outlines a commendable national vision but, as the report notes, “lacks explicit budget lines or projections, relying on external partners.” 

In 2024, UNDP, UNESCO, Meta, Google, and Microsoft jointly provided $3.5 million in seed funding to jump-start implementation.

There is progress; the N-ATLAS model now supports Yoruba, Hausa, and Igbo; the AI Scaling Hub, a Gates Foundation partnership, is expanding use in health, education, and agriculture.

Still, “the strategy also has no binding legislation; it remains a policy rather than law.” 

It is this mixture of visible innovation and fragile governance that defines both Nigeria and the continent’s AI story.

A Continental Picture of Uneven Progress

Across 20 African countries, more than half have drafted or adopted national AI strategies. However, fewer than one in four have committed budgets, enforceable laws, or monitoring systems.

AI, the report argues, has become both a symbol of modernity and a test of governance capacity.

The AI Governance Maturity Index rates countries out of 24 points across four categories: policy design, implementation capacity, governance and accountability, and external engagement.

Below these leaders, Namibia, Botswana, and Zimbabwe remain in early drafting. 

A further 34 countries, including Chad, Sierra Leone, and Eswatini, have no AI policy at all.

Follow the Money

Funding exposes the depth of the gap. Kenya’s $1.1 billion allocation dwarfs Zambia’s U.S. $335,000 budget by more than 300 times.

Ethiopia increased its national AI budget by 42% in 2025. Egypt, the regional heavyweight, continues to pull foreign capital through a €1.8 billion hyperscale data centre and a $300 million semiconductor investment.

Yet most countries depend on donors.

“Only a few governments have dedicated, multi-year AI budgets; most depend on donor support or general ICT allocations that are easily diverted,” the report cautions.

Where Law and Ethics Lag

Legal infrastructure is also far behind rhetoric. While 35 African nations now have data-protection statutes, almost none have AI-specific regulation. 

Egypt’s framework remains voluntary; Ethiopia’s ethics guidelines are not binding; South Africa’s draft bill awaits ratification. Nigeria, the authors note, operates on policy intent rather than legal force.

The result is what Column calls “ethics without accountability,” a moral vocabulary without enforcement.

The Transparency Deficit

Few governments publish metrics or progress reports. Egypt stands out for tracking outcomes through measurable key performance indicators. 

Elsewhere, “too many strategies are unpublished or inaccessible,” the report says, “reducing transparency and accountability.”

This opacity makes it hard for citizens, researchers, or investors to know whether AI spending yields tangible results.

Regional Contrasts

• North Africa (Egypt, Algeria): strong institutions, industrial investment.
• East Africa (Ethiopia, Kenya, Rwanda): innovation and experimentation.
• West Africa (Nigeria, Ghana, Senegal, Côte d’Ivoire): numerous strategies, weak enforcement.
• Southern Africa (South Africa, Zambia, Lesotho): policy structure, modest budgets.

The data reveal a continent moving at different speeds toward the same uncertain finish line.

Why It is Important 

AI is not only a technological issue but also a governance test. It could improve crop yields, detect disease, streamline justice systems, and expand financial inclusion.

However, as the authors warn, “Without strong governance, it can deepen inequality, embed bias, or be used for surveillance and censorship.”

With more than 60% of Africans under 25, the economic stakes are immense. Countries that craft and enforce sound AI policies will shape not just algorithms but their own development futures.

“Countries that shape the technology also shape the rules,” the report reminds readers.

From Paper to Proof

To bridge the gap between ambition and delivery, the authors urge African governments to:

1. Legislate AI frameworks rather than rely on non-binding strategies.
2. Fund multi-year national programmes.
3. Establish independent bodies for ethics and accountability.
4. Publish monitoring data for public scrutiny.

“Africa is not behind—it is early. The task now is to make ambition durable: to move from promise to proof.”

Nigeria as Mirror and Test Case

In many ways, Nigeria encapsulates Africa’s AI journey, a nation bursting with innovation, global partnerships, and youthful expertise, yet constrained by governance policy and finance. 

Its N-ATLAS language model and AI Scaling Hub demonstrate what is possible when local capacity meets global collaboration. But without statutory backing or sustainable funding, such progress risks fading into headlines rather than history.

As the State of AI Policy in Africa 2025 makes clear, the next frontier will not be who drafts the most visionary strategy, but who brings measurable, lawful, and lasting impact.

A new report by Column and Smarketers Hub reveals that while 71% of marketing teams across Africa already use AI tools regularly, only 26% have received formal training. 

Hence, Marketers are saving up to 10 hours a week with tools like ChatGPT, yet stumbling when it comes to strategy, automation, and data analysis.

The survey of 100 marketers, 90 based in Nigeria and others from Ghana, Zambia, and beyond, reveals that AI has become a fixture in everyday marketing. 95% of respondents use ChatGPT, 55% rely on Gemini, and 42% on Claude. Smaller groups use tools like Copy.ai (23%), Jasper (9%), DALL·E (11%), and Midjourney (4%). 

For 82% of marketers, content creation is the top use case, followed by ad copy (25%), audience research (22%), SEO (18%), and reporting (14%).

And the benefits are undeniable. 41% of respondents reported saving 4 to 6 hours per week, while 18% said AI gives them back more than 10 hours. Freelancers, who made up 15% of respondents, described AI as a lifeline, filling the gaps in teams too small to meet deadlines alone. 

One respondent said: “AI gives me a good structure. I then put my voice in the work before sending it out.”

But then, 57% described AI’s impact as very positive, 30% as somewhat positive, while only 8% held a neutral or negative view. When asked what still gets done manually, the answers were: “Everything.” and “99% of my work is content writing. So nothing I do is completely automated. AI-assisted? Yes. But it’s all manual.”

The problem isn’t desire but direction. 46% of marketers feel least confident in technical SEO, 40% in workflow integration, 39% in data analysis, and 38% in automation. 

Limitations are familiar, as 31% don’t know where to start, 26% lack time, 19% cite poor training resources, and 18% blame tight budgets. Despite these challenges, companies are offering little help, most respondents said they had received no AI-related training in the past year.

Aisha Owolabi, founder of Smarketers Hub, said “This report is a wake-up call for marketing leaders in Africa: AI isn’t just a future trend, it’s already reshaping how teams work. The data makes it clear — marketers are eager and experimenting, but without structured support they’ll remain stuck at the basics.”

Mo Shehu, CEO of Column, reiterated the urgency: “AI is reshaping how African marketers work. Teams are experimenting even without much formal training and already saving hours each week. The challenge for CMOs is turning those small wins into a structured, team-wide advantage.”

The report classifies teams into three categories. Beginners tinker without guidance, producing scattered results. Intermediates experiment more actively, especially in content and SEO, but lack standards and consistency. 

Only a handful of advanced teams embed AI fully into workflows, supported by playbooks, ongoing training, and ROI tracking; most African marketers today sit at the beginner or early-intermediate stage.

Looking deeper into the demographics, we see 88% of respondents were junior or mid-level marketers, meaning the future of AI in African marketing rests on early-career professionals with limited resources. For them, structured mentorship, localised training, and practical playbooks are highly important.

The report further forecasts three changes in the next 12 months:

1. Tool diversification: As Marketers get more confidence, they will start moving beyond ChatGPT into tools like Gemini, Claude, Canva AI, and Perplexity.
2. Formal training: companies embedding AI modules into onboarding, workshops, and performance reviews.
3. Team-level alignment: AI shifting from individual experimentation to coordinated campaign planning, reporting, and strategy.

But risks are increasing too. Regulation of AI use, particularly around data, content transparency, and ethics, is expected to tighten, both from governments and companies. The report warns CMOs to establish internal codes of conduct now, before misuse catches up with them.

“African CMOs have a chance to lead, not follow,” Owolabi said. For now, the continent’s marketers are eager but underprepared, productive but undertrained, saving hours every week but losing years of competitive advantage without the structure to take AI beyond the basics.

Micro, Small, and Medium-sized Enterprises (MSMEs) don’t need to spend millions to benefit from AI. Nigeria is home to 40 million MSMEs, contributing 90% of employment. MSMEs also account for 96.9% of businesses, 87.9% of employment, and contribute 46.3% of GDP. That means nearly every job or service around you has MSMEs at its core.

Despite their importance, most MSMEs still struggle with high inflation, unstable electricity, and rising costs. A July 2024 PwC survey of 567 MSMEs found that 67% reported declining customer demand due to reduced purchasing power.

In this kind of climate, the pressure to cut costs while staying productive is higher than ever. This is exactly where AI can come in.

Chances are, you’ve already used AI without realizing it. Voice assistants like Siri and Alexa run on AI, as do the chatbots that pop up when you’re browsing a website, trying to find your way around.

Many business owners still think AI is for the future. Others fear it will replace jobs, but the reality is, AI at the MSME level is not about robots taking over your bakery or replacing your shop assistant. It’s about helping you work smarter, cut waste, and serve customers better.

Say you run a small fashion brand. Customers keep messaging you on WhatsApp to ask the same questions: “How much is this dress?”, “Do you deliver to Abuja?”, “What’s your size guide?”

Instead of answering every single one manually, you can set up a simple WhatsApp chatbot using platforms like BotMe or Wati. These tools can answer FAQs 24/7, take orders, and even follow up with customers politely.

Or perhaps you’re in the food services industry, struggling to keep track of expenses. Instead of jotting everything down in a notebook, you could try AI-driven bookkeeping tools like Tyms AI or Vic.AI. These apps can automatically record sales, generate invoices, and remind you of pending payments.

And Nigeria is more digitally connected than ever before. The Nigerian Communications Commission (NCC) reports that as of June 2025, there are 140.6 million active internet subscribers and 48.7% broadband penetration. That’s almost half the country online.

In fact, total data consumption has shot up to 1.04 Terabytes per month across nearly 171.7 million Nigerians. So if you think your customers aren’t online, think again. They’re browsing Instagram, searching Google, and scrolling TikTok. With AI, your business can meet them right there.

Let’s also talk about marketing. Instead of paying a marketing agency hundreds of thousands of naira, you can use ChatGPT, Claude.AI, or Google Gemini to draft social media captions, email campaigns, or product descriptions.

If you need visuals, tools like Canva AI and Freepik can generate designs in minutes. Use Descript for audio and Google Veo3 for video. Suddenly, the marketing team you thought you couldn’t afford is right at your fingertips.

These are not distant or futuristic dreams. They are practical shifts already happening around you, available at a fraction of the cost once imagined.

Of course, barriers exist. Many MSMEs fear that AI is too expensive, while others struggle with poor digital literacy or unreliable internet. But think about how people adapted to mobile banking. At first, USSD codes seemed confusing. But today, even market women dial *737# – a USSD shortcode provided by Guaranty Trust Bank (GTBank) for simple banking services on any mobile phone, including transfers, airtime purchases, bill payments, and balance inquiries – without blinking. The same can happen with AI: once we see the value, adoption will spread naturally.

Mindset change is key. If your MSME continues to run its business the old way, you’ll keep losing customers to those who can serve faster, cheaper, and better.

Policymakers also need to wake up. McKinsey estimates that Generative AI alone could unlock up to $100 billion in annual economic value across Africa. Supporting low-cost AI adoption through tax breaks, training programs, and digital infrastructure is not just a nice idea, but a national urgency.

So, where to begin? Start small. Try a WhatsApp chatbot to onboard new users or answer FAQs. Use ChatGPT to write marketing content. Replace manual receipts with AI invoicing tools. Even if you just start with one tool, you’ll quickly see the time and money saved.

============================================================

*Mo Shehu, PhD, is the CEO of Column – a UK-based media and research firm. With over a decade of experience across three continents, Dr. Shehu helps leaders and brands grow through clear, credible insights. He lives in the UK.