Ugandan tech and business leaders speak out on what is working and what CEOs should take from it
Over the last two to three years, artificial intelligence has dominated corporate conversations in Uganda.
It has been debated in boardrooms, unpacked at conferences, written about extensively and weighed in strategy sessions. What it can do, what it cannot do, and whether it makes sense for local organisations to invest at all.
For many institutions, those discussions led to caution rather than action, as leaders waited for clearer rules, proven models or perfect conditions.
Yet even as some organisations hesitated, others quietly moved ahead. Individuals and early adopters began experimenting with AI tools, embedding them into everyday workflows and using them to solve practical problems.
From fraud detection and credit assessment to customer service and productivity. Today, the impact of those early decisions is becoming visible.
AI is no longer confined to innovation labs or pilot projects; it is increasingly embedded in the operational backbone of banks, technology firms, real estate companies and parts of government.
As a result, the nature of the debate has shifted. The question is no longer whether AI matters. That has been settled. The more urgent issue now confronting Ugandan executives is how far AI can be taken, and how deeply it can be integrated into core operations.
How responsibly it can be governed, and how quickly organisations can build the skills, data discipline and infrastructure needed to scale its use. Execution, not aspiration, is now the true test of AI readiness.
In this special report, CEO East Africa Magazine speaks to Ugandan tech and business leaders who have moved beyond theory into practice. They share what is working, where the constraints remain, and why execution, rather than aspiration is now the true measure of AI readiness.
That shift from debate to execution is now visible across Uganda’s corporate landscape. AI is no longer confined to conference stages or innovation labs.
It is increasingly embedded in boardroom decision-making, bank risk teams, customer service channels and the less glamorous but critical operational plumbing of data-driven businesses.
Yet this transition is unfolding unevenly. While the opportunity is significant, so too are the structural constraints, particularly around data readiness, skills, governance frameworks, trust and access to affordable computing power.
What emerges from voices across banking, technology and real estate is a clear and consistent message: AI is not magic, nor is it a plug-and-play solution. It is an operating capability.
Organisations that treat it as such, deploying it with intent, discipline and clearly defined use cases—are beginning to pull ahead.
Those who approach AI as a side experiment or who wait for perfect conditions risk discovering that the execution gap is widening faster than anticipated.
The digital, always-on customer
For Steven Mwesige, Acting Chief Information Security Officer at Pearl Bank Uganda Limited, the push towards AI begins with a simple reality: banking is now deeply digital and intensely data-driven.
“Customers want real-time responses, personalised service, seamless onboarding, quicker loan decisions and support at any hour.
At the same time, banks must process and interpret vast volumes of information generated across mobile apps, USSD channels, agency networks and interoperability platforms,” he shares.
In Uganda and the wider East African region, this shift is amplified by high mobile money usage and rapidly expanding digital transactions. The result, Mwesige argues, is that AI is increasingly important not because it is fashionable, but because it is practical.
That is by helping banks make faster decisions, improve customer experience, and manage risk in an environment where threats are rising.
Fraud, cyber risk, speed and credit inclusion
As digital traffic grows, so does criminal creativity. Mwesige highlights fraud detection and cybersecurity as some of AI’s most immediately transformative use cases for the region.
Behavioural analytics can learn normal customer patterns and identify anomalies across channels, often faster and more accurately than manual monitoring. That matters in markets where fraud attempts evolve quickly, and attacks increasingly rely on manipulation and social engineering, not just technical exploits.
The goal is not merely to “catch fraud” after the act, but to detect it earlier, reduce losses, and protect confidence in digital financial services; confidence that underpins the entire ecosystem.
While AI addresses practical local challenges that ease bank operations in a rapidly digitising regional economy, it also reduces fraud, which many banks are committed to keeping low. The double-edged applicability of AI keeps people on their toes.
Beyond security, Mwesige identifies AI-driven alternative credit scoring as one of Uganda’s biggest opportunities. Many customers, especially MSMEs and informal-sector earners, lack formal credit histories.
Yet they may have rich behavioural trails: mobile usage patterns, payment behaviour, transaction consistency and other digital cues.
If used responsibly, AI could help banks make more accurate lending decisions and expand access to credit, particularly for those traditionally excluded by conventional scoring.
In a country where financial inclusion remains a national priority, this is one of the most compelling “local fit” arguments for AI.
“Operationally, intelligent automation can streamline reconciliations, compliance checks, and reporting critical areas for banks, balancing cost pressures and regulatory requirements.
These use cases represent both high impact and high relevance for the region’s financial ecosystem,” he says.
AI as a personal assistant, not a replacement
From the builder’s perspective, Allan Rwakatungu, CEO of Intelligent Tyms, sees AI’s most immediate value differently: as assistance.
In his view, the breakthrough is that “every human can finally have a smart personal assistant”, a tool that helps answer questions, complete tasks, and understand context, both personally and at work.
“This is not about replacing staff, but lifting productivity and transforming what an individual can accomplish in a day,” he says.
That framing is important in Uganda’s labour context: if adoption is positioned as augmentation rather than displacement, it becomes easier to build internal confidence and accelerate use.
A boardroom issue
Rwakatungu is emphatic that AI cannot be delegated to the IT department as a side project. He says AI is a strategy, a boardroom issue. Moreover, he says that we do not have to follow Silicon Valley’s playbook.
“We can skip straight to AI-powered operations built for our context, our challenges, our opportunities,” he says.
He describes it as a leapfrog moment. A chance to rethink operations rather than bolt technology onto old processes.
The organisations that win will not be the ones with the fanciest pilot, but the ones with leadership willing to decide, invest, learn quickly and execute.
His biggest concern is not simply governance or cost, but hesitation: whether businesses will wait for perfect conditions while others move, iterate and compound their advantage.
“Our concern is the market. Is it ready? Can it keep up with what’s possible? Will we miss the boat while others move?” he ponders.
The unglamorous foundation of governance
If speed and ambition pull organisations forward, governance keeps them safe.
Moses Lutalo, Managing Director at Broll Uganda, says AI should enhance professional judgement, not replace it, particularly in sectors such as real estate, where recommendations and valuations have significant consequences.
Before using AI at scale, he argues, firms need strong foundations. These include clear data governance, bias and fairness checks, and privacy and consent controls. That is not to forget the auditability of AI-generated insights and the need for human oversight of outputs.
“This is not bureaucracy for its own sake. It is the difference between AI that improves decisions and AI that quietly introduces risk,” he says.
But he also agrees with Rwakatungu that AI should enhance professional judgement, not replace it.
Government readiness
National Information Technology Authority – Uganda (NITA-U) recognises significant progress in Uganda’s readiness for advanced digital technologies, driven by initiatives such as the Uganda Digital Acceleration Project (UDAP).
These have enabled the adoption of these technologies. Beatrice Nabukenya, a Cloud Services Engineer at NITA-U, says efforts to strengthen and expand fibre backbone connectivity and secure data hosting showcase have improved digital accessibility.
“Uganda is also advancing digital identity systems such as UGPass, laying the groundwork for secure, interoperable digital services. Many government agencies already use cloud services, and these partnerships aim to enhance e-government,” she says.
Related
Serigne Dioum: Powering a $500 Billion Fintech Engine at the Heart of Africa’s Digital EconomyAs technologies become more accessible, cybersecurity remains a priority, with coordinated national incident response under UgCERT and improved resilience frameworks across the government and private sectors.
The constraint of data quality
Lutalo’s view on real estate readiness in Uganda is candid: the sector remains early-stage.
Many firms operate with fragmented datasets, unstructured records, manual processes, and limited cross-departmental integration. His message is blunt but realistic: AI cannot outperform the quality of its inputs.
This is something close to Mustapha Mugisa’s heart because he reminds us that without data, there is no AI.
The route to progress is therefore practical, not theoretical. It is clean, standardised data; modern property management systems; better documentation habits; stronger cybersecurity.
Eventually, sector-wide standards will enable data to be used more consistently.
The glaring hurdles and AI-induced problems
Mugisa, the Director of the Institute of Forensics and ICT Security, adds a more sceptical but important counter-weight: many organisations in Uganda are not truly “adopting AI” end-to-end, at least not yet.
He says that most companies are using basic generative AI in small parts of their work, often by connecting to external tools rather than building robust internal systems. The main barrier, in his view, is computing power.
That is because AI at scale requires substantial infrastructure, and the costs are difficult for individual firms to bear.
“Uganda does not have the energy supply to afford the consumption for an AI data centre, which requires huge computing power,” he says.
His wider point is that national readiness matters too. That is energy reliability, data centre capacity, cloud affordability and policy direction to influence whether AI remains superficial or becomes embedded.
Mugisa also highlights a subtle yet critical safety concern: bias. For instance, poorly designed AI systems can embed unfair rules that exclude entire groups from services, especially in financial decision-making.
That risk increases when organisations lack well-structured local data and rely on global public datasets that may not reflect Ugandan realities.
He says a minimum viable governance approach must include ethics, clear codes of conduct, and safeguards that prevent discriminatory outcomes without ‘killing innovation’.
“This governance must also not pretend that risk will solve itself,” he says.
Change management: Winning hearts before automating workflows
A recurring theme across voices is that adoption is as much cultural as technical.
Boards and staff often fear automation because they associate it with job losses, loss of control, or reputational risk. Mugisa’s practical answer is education and staged adoption.
“Take the leadership and teams through AI literacy to build confidence through small wins, and introduce AI in ways that help staff do their jobs better. That must happen before pushing deeper automation,” he advises.
This aligns with Mwesige’s workforce view: banks need structured readiness programmes, cross-functional collaboration, continuous training (not one-off workshops), and a culture that rewards experimentation.
Continuous skilling and appraisals are important as organisations embrace AI because the technology is continually evolving.
Role of the government
Nabukenya agrees with Mugisa that, without stifling private-sector agility, the government should set clear, risk-based guidelines that protect privacy, uphold ethical standards, and foster trust, while avoiding biased controls that slow innovation.
“This involves enforcing the Data Protection and Privacy Act, conducting cybersecurity and impact assessments for sandboxes where start-ups experiment with AI solutions safely,” she says.
The state should also incentivise public-private partnerships and align national research and innovation funds with industry needs to create value from innovation.
By balancing stewardship with flexibility, Nabukenya says government can ensure AI benefits citizens without dampening private-sector creativity and competitiveness.
Lastly, collaborative public-private investment in digital infrastructure and skills development will provide the needed skills for AI innovation.
What can the CEO do?
Lutalo says skills can be developed, but leadership must drive the culture shift. Therefore, he advises the CEO to start small and build momentum.
That could take the form of identifying internal staff with an aptitude for analytics, training them on AI tools, and then pairing them with external experts for the first 6–12 months.
“Thereafter, create mini-AI labs or innovation squads headed by these people. Then reward teams that automate manual processes,” he says.
With that, a real estate company can benefit from predictive maintenance (reducing breakdowns & Opex), automated service-charge reconciliation, and data-driven leasing insights.
That includes tenant enquiry chatbots, faster, more consistent property valuations, and portfolio performance dashboards.
“These deliver savings, speed, and better decision-making,” he says.
Collaboration
Clearly, for AI to thrive and benefit all players, there must be a collaboration between the government and the private sector.
Nabukenya says that while a comprehensive AI framework is being developed, Uganda is advancing sector-specific considerations through existing regulatory frameworks and digital strategies.
This is in areas such as the judiciary, health, finance, and telecom sectors, which operate under robust risk and compliance regimes. That is particularly around data security and consumer protection.
In healthcare, Nabukenya says the discussions emphasise ethical use of data and the integration of AI into health information systems to support Universal Health Coverage.
“Sector regulators are increasingly expected to issue tailored standards and guidelines aligned with national digital transformation and data governance goals, ensuring responsible AI adoption across critical industries,” she says.
Who thrives and who falls behind?
The dividing line is becoming clearer.
Mwesige argues that thriving banks will treat AI as a strategic capability, invest in modern infrastructure, build strong governance, and embed AI into customer experience, risk management, and decision-making.
Rwakatungu reinforces the leadership angle: the winners will be agile organisations that act quickly, learn quickly, and remain market- and customer-centric.
Lutalo adds the operational reminder: none of this scales without data discipline, cybersecurity and human oversight, especially in sectors where trust is the product.
The bottom line
Uganda does not need to copy another region’s playbook. But it needs to move from conversation to implementation.
AI’s promise is clearest where it meets Uganda’s real conditions: mobile-first behaviour, high transaction volumes, fraud pressure, inclusion needs, and the productivity challenge across industries.
Yet the biggest constraint is equally real: without cleaner data, stronger governance, better skills, and infrastructure capable of supporting serious workloads, AI will remain shallow.
The next chapter, therefore, will not be defined by who talks about AI the most. It will be by those who quietly build the foundations, then use AI to deliver faster decisions, safer systems, better service and measurable value.
But in all this, Mugisa says the country must be ready before companies within its borders, because real business happens when infrastructure is in place and operational.
“Otherwise, the costs are so astronomical to be borne by a few people,” he says.
It will also help if the government looks beyond AI’s risks to drive its continued growth in Uganda.
