Artificial intelligence (AI) is increasingly changing how businesses use technology, but its wider impact may be less about replacing existing systems than about forcing organisations to modernise the data, infrastructure, skills and governance on which those systems depend. The shift is becoming particularly relevant for African businesses, where uneven digital adoption, skills shortages and fragmented technology infrastructure could determine how quickly companies capture productivity gains from AI while managing its energy, cybersecurity and governance costs.
The question of whether AI will change technology or technology will change AI captures a feedback loop now developing across enterprise systems. Advances in AI are creating new applications and automated workflows, while improvements in cloud computing, enterprise software, data architecture and connectivity are expanding what AI systems can do. The result is a technology cycle in which each development increases demand for the other.
A recent SAP Community article framed the issue through an everyday conversation between a technology user and a hairstylist, using the evolution of calculators and the internet to illustrate how technologies initially viewed with caution can eventually change the nature of work rather than simply eliminate it. The comparison is relevant to the current AI debate because the economic effects of the technology are likely to depend heavily on how organisations redesign work around it.
Calculators did not remove mathematics from education, while the internet did not simply replace existing information systems. Both changed how people performed tasks and shifted attention towards activities that required interpretation, judgement and problem-solving. AI is now producing a similar adjustment across business functions, automating activities such as document summarisation, data reconciliation, information retrieval and routine analysis.
In procurement, for example, AI systems can increasingly support contract drafting, supplier assessment, sourcing, purchasing and risk analysis. SAP’s recent enterprise applications illustrate how AI can combine information from supplier records, contracts, financial indicators and ESG data to give procurement teams a more integrated view of suppliers and purchasing decisions.
The implications extend beyond procurement. Finance, logistics, human resources, customer service and supply-chain management are becoming potential areas for AI-assisted decision-making. But the ability to deploy these systems effectively depends on the quality and accessibility of the data they use.
That requirement presents a particular challenge for African enterprises. Many businesses are still operating with fragmented systems, limited connectivity, inconsistent data practices and shortages of specialised digital skills. The result is a gap between access to AI tools and the ability to deploy them at enterprise scale.
Recent research cited by SAP Africa illustrates the scale of the skills challenge. African organisations report rising demand for capabilities in artificial intelligence, cybersecurity, cloud computing and data analytics, while shortages in these areas are already affecting implementation and innovation. SAP’s 2025 reporting on AI skills in Africa found that almost 90% of surveyed organisations reported consequences from AI skills shortages, including implementation delays, failed innovation initiatives and an inability to take on new work.
For businesses, this means that AI adoption cannot be treated as the procurement of another software product. It requires investment in the systems surrounding the technology. Companies need reliable data, modern enterprise platforms, cybersecurity controls, clear decision rights and employees capable of working with AI-generated outputs.
This is becoming particularly important as enterprise AI moves from tools that assist employees towards systems capable of taking actions within defined business processes. SAP’s African market reporting has highlighted the growing use of agentic AI, where systems can execute multi-step tasks rather than simply generate recommendations.
That shift changes the governance question. A system that produces a draft report can be reviewed before it is used. A system that automatically initiates a procurement process, changes a workflow or communicates with a supplier introduces a different level of operational risk. Organisations therefore need clear boundaries around what AI can decide, what requires human approval and how decisions can be audited.
For African regulators and businesses, the issue is not only technological. Data protection, cybersecurity, intellectual property, employment and accountability frameworks will increasingly intersect with AI adoption. The more autonomous these systems become, the more important it becomes for organisations to know what data they are using, where that data is stored, how models reach conclusions and who remains responsible when an automated decision causes financial or operational harm.
The sustainability implications are also becoming harder to separate from the technology discussion. AI systems depend on data centres, computing infrastructure and electricity. As organisations expand their use of AI, demand for computing capacity could increase the pressure on already constrained energy systems, particularly in markets where electricity reliability remains a challenge.
For African economies seeking to expand digital infrastructure, this creates a policy trade-off. Investment in data centres, cloud infrastructure and telecommunications can support productivity and digital services, but those facilities also require reliable electricity, cooling systems and physical infrastructure. The environmental footprint of digital expansion will therefore become increasingly connected to national energy planning.
This is especially relevant as African countries pursue both digitalisation and energy-transition strategies. The growth of AI could increase demand for renewable-powered data infrastructure, efficient cooling technologies and better energy management. In that context, digital transformation and sustainability may increasingly become linked investment decisions rather than separate policy areas.
The same relationship applies to supply chains. AI can improve the ability of companies to identify supplier risks, analyse purchasing patterns and respond to disruptions. Better data can also help businesses track environmental and social information across their supply chains, an increasingly important requirement as sustainability disclosure and responsible sourcing standards expand.
For African exporters, this could become commercially significant. Companies supplying international markets are increasingly expected to provide information on emissions, resource use, labour practices and supply-chain risks. Digital systems capable of collecting and analysing this information can help businesses respond to those requirements, while poorly integrated systems may increase the cost of compliance.
AI could therefore become part of the infrastructure supporting ESG reporting and sustainable procurement. But its usefulness will depend on the quality of the information feeding into it. An advanced AI system cannot compensate for inaccurate supplier records, incomplete emissions data or poorly governed corporate information.
The same principle applies to decision-making. Faster analysis does not necessarily produce better decisions. AI can identify patterns and generate recommendations at a speed that humans cannot match, but organisations still need people capable of evaluating whether those recommendations make commercial, ethical and regulatory sense.
This is particularly important in African markets where business decisions can involve complex informal supply chains, regulatory environments and infrastructure constraints that may not be adequately represented in standard datasets. Human judgement remains important because the data available to an AI system may not fully capture the context in which a decision is being made.
The skills transition is consequently broader than training employees to use AI prompts. Businesses will increasingly need workers who understand data, cybersecurity, process design, risk management and the limitations of automated systems. At the same time, workers whose roles involve repetitive administrative tasks may need opportunities to develop higher-value analytical and decision-making capabilities.
That shift could have significant implications for Africa’s labour markets. The continent has a large and growing working-age population, but many economies already face difficulties matching education and training systems with employer demand. AI could intensify that mismatch if companies rapidly adopt new technologies without investing in workforce development.
There is evidence that African businesses are already moving towards a skills-based approach to technology hiring. SAP Africa reported in March that organisations are increasingly focusing on specific capabilities in areas such as AI, cybersecurity, cloud computing and data analytics, while warning that foundational training, mentorship and practical exposure remain uneven.
For governments, the challenge extends beyond corporate training. Digital infrastructure, affordable connectivity, reliable electricity and access to computing resources will influence which businesses can benefit from AI. Large companies with stronger balance sheets may be able to invest in cloud systems and specialised talent, while smaller enterprises could face higher barriers to adoption.
That matters because small and medium-sized enterprises account for a large share of African economic activity. SAP Africa estimates that SMEs represent roughly 95% of registered businesses in Sub-Saharan Africa and around half of regional GDP, while World Bank research has found that fewer than one in three African firms using digital technologies make intensive use of them to improve business operations.
The risk is that AI could widen an existing productivity divide if access to advanced technology becomes concentrated among larger firms and digitally mature economies. The alternative is for governments, financial institutions and technology providers to focus on the foundational conditions that allow smaller businesses to adopt digital tools without having to build complex technology infrastructure from scratch.
Cloud computing and shared digital platforms could reduce some of those barriers. But adoption still requires financing, skills and trust. Businesses need confidence that their data is secure, that systems comply with applicable regulations and that technology investments will produce measurable commercial returns.
For corporate leaders, this changes the question from whether to adopt AI to where AI can create measurable value. SAP’s recent reporting from Africa has similarly emphasised a shift from experimentation towards enterprise-scale applications with defined business outcomes.
The distinction is important because AI investment can become expensive when organisations adopt technology without redesigning the processes around it. A business may deploy an AI tool while maintaining outdated approval structures, fragmented databases and manual workflows, limiting the value of the technology.
The experience of Standard Bank provides an example of the infrastructure challenge. The bank has been modernising its enterprise systems as part of a wider technology transformation, with executives identifying data, cybersecurity, workforce readiness and change management as central considerations in preparing for further AI adoption.
This suggests that the most important AI investments may not always be the most visible. Data architecture, cybersecurity, cloud migration, system integration and workforce development are less conspicuous than generative AI demonstrations, but they determine whether organisations can safely scale the technology.
For Africa, that distinction has wider economic implications. The continent’s digital transformation is occurring alongside pressure to improve productivity, create jobs, strengthen supply chains and reduce the resource intensity of economic growth. AI could contribute to those objectives, but only if digital investment is connected to broader infrastructure and human-capital strategies.
The sustainability question will also become more prominent as AI infrastructure expands. Countries seeking to attract data centres and digital investment will need to consider electricity availability, renewable-energy capacity, water requirements and network infrastructure alongside tax incentives and technology policies. For economies already managing electricity shortages or water stress, the resource implications of large-scale computing infrastructure cannot be treated as secondary considerations.
At the corporate level, AI could also support sustainability performance by improving energy management, identifying supply-chain inefficiencies and analysing environmental data. The potential benefits, however, depend on whether organisations integrate AI into operational decision-making rather than treating it as a separate innovation programme.
The broader lesson is that AI and technology are likely to continue reshaping each other. The direction of that change will be influenced not only by advances in algorithms but also by the infrastructure, markets, regulations and skills surrounding them.
For African businesses, the immediate challenge is therefore one of readiness. Companies that modernise their data and operating systems, invest in relevant skills and establish clear governance structures may be better positioned to use AI as an operational tool rather than simply as an experimental technology.
For policymakers, the challenge is broader: ensuring that AI investment contributes to productive capacity without deepening digital inequality, increasing infrastructure pressures or creating governance gaps. That requires digital policies to be considered alongside energy, education, cybersecurity, industrial development and sustainability strategies.
The question raised by a conversation in a salon therefore points to a much larger economic issue. AI is changing technology, but technology infrastructure is simultaneously determining how far AI can go. In Africa, the outcome will depend less on the availability of AI tools than on whether businesses and governments can build the systems, skills and institutions needed to use them productively, securely and sustainably.
