AI Could Affect 40% of Global Jobs: What Businesses Must Do to Prepare

The IMF’s estimate points to a major shift in how work is performed. For businesses, the challenge is to capture productivity gains while preparing employees for changing roles and skill requirements.

Artificial intelligence is becoming a business issue that extends far beyond technology departments. It is changing how companies approach customer service, research, marketing, software development, data analysis and routine administrative work and raising new questions about hiring, training and the future shape of the workforce.

The International Monetary Fund estimates that almost 40% of global employment is exposed to artificial intelligence. The finding, published in its research on generative AI and the future of work, highlights the potential scale of the technology’s influence on labour markets.

But the figure requires an important qualification: exposure to AI does not mean a job will necessarily disappear. Some roles could be transformed or partly automated, while others could become more productive when employees work with AI tools. The eventual outcome will depend on the tasks involved, how employers deploy the technology and whether workers can adapt.

For business leaders, the central question is therefore not simply how many jobs AI might replace. It is which activities can be improved, which capabilities employees will need and how companies can introduce the technology without undermining performance, security or customer trust.

What the IMF’s 40% estimate actually means

The IMF’s research distinguishes between exposure to AI and the likely relationship between AI and the tasks people perform.

Some jobs contain tasks that AI systems may be able to perform or support. In certain roles, this could allow employees to complete work faster, improve analysis or devote more time to activities that require human judgement. In others, AI could perform tasks that previously required paid human labour, potentially reducing demand for particular roles.

The IMF’s analysis estimated that around 60% of jobs in advanced economies are exposed to AI, compared with approximately 40% in emerging-market economies and 26% in low-income countries.

These differences partly reflect the kinds of work performed in different economies and their readiness to adopt the technology. Economies with a larger share of cognitive-intensive jobs may experience greater exposure. But having less immediate exposure does not automatically put a country in a stronger position: weaker digital infrastructure, limited access to skills and lower adoption capacity can also make it harder to capture AI’s productivity benefits.

The figures should not be interpreted as a forecast that 40% of jobs globally, or 26% in low-income economies, will be eliminated. They describe potential exposure to AI-related change, not a guaranteed outcome for employment.

Why this matters to business owners

For businesses, AI can influence costs, productivity, competition and the skills required to operate effectively.

A company that automates parts of its administrative work may be able to process more transactions without increasing headcount at the same rate. A customer-service team might use AI to classify enquiries and draft routine responses, leaving employees to handle complex complaints. An analyst may use AI to organise information faster but still need to verify the data and interpret the results.

These are examples of possible applications, not guaranteed savings. The business benefit depends on whether the technology works reliably, fits the task and costs less than the value it creates.

Adoption also creates expenses and risks. Companies may need new software, staff training, data protection measures, cybersecurity controls and human review processes. Poorly implemented AI can introduce errors, expose confidential information or generate work that requires expensive correction.

The relevant measure is not how many AI tools a business has adopted. It is whether those tools improve a clearly identified business outcome.

Which business functions could change?

Customer service and sales: AI can help organise enquiries, draft responses, summarise customer interactions and support sales teams with research. Human employees remain important for sensitive complaints, negotiation and relationship management.

Finance and administration: Businesses may use AI to assist with document processing, invoice categorisation, report preparation and internal information retrieval. Financial outputs still require appropriate checks, particularly where errors could affect payments, reporting or compliance.

Marketing and communications: AI can help generate initial drafts, analyse campaign performance and adapt content for different audiences. Businesses still need people to check factual accuracy, protect brand reputation and understand local customers.

Research and analysis: AI can accelerate the organisation and summarisation of large amounts of information. But an AI-generated summary is not proof that a source is accurate, and important decisions should be based on verified data.

Technology and operations: AI-assisted coding, troubleshooting and process monitoring can change how technical teams work. Employers must assess output quality, system security and the consequences of errors before relying on automation for critical processes.

The effect will vary by company, industry and role. Businesses should assess individual tasks rather than assume that an entire occupation can be automated simply because some of its work is compatible with AI.

The risk of waiting too long to prepare

Companies that delay all AI-related decisions may miss opportunities to improve productivity, shorten turnaround times or serve customers more effectively. Yet adopting the technology without a clear purpose can waste money and create operational problems.

There is also a workforce risk. If businesses change job requirements faster than employees can acquire new skills, they may face capability gaps, lower morale and difficulty recruiting for emerging responsibilities.

The IMF has highlighted the importance of adapting skills as AI reshapes work. Its January 2026 analysis reported that around one in ten job postings in advanced economies and one in twenty in emerging-market economies required at least one new skill.

That finding reinforces a practical point for employers: workforce development cannot be treated as an afterthought. Companies need to identify which skills are becoming more valuable and give employees a realistic path to acquire them.

Five steps businesses can take now

1. Audit tasks before cutting roles.

Identify repetitive, time-consuming or information-heavy activities that may benefit from AI support. Separate those tasks from responsibilities that require judgement, accountability, negotiation, creativity or direct customer relationships. Do not assume that automating one task makes an entire position unnecessary.

2. Test AI against a measurable business problem.

Choose a limited use case, such as reducing the time needed to process documents or helping staff find internal information. Establish a baseline, run a controlled pilot and measure accuracy, time saved, cost, customer outcomes and the amount of human correction required.

3. Train employees alongside adoption.

Provide practical training in the tools employees are expected to use. Training should include how to check outputs, recognise limitations, protect sensitive information and escalate problems. Managers also need to understand how AI changes workflow and performance expectations.

4. Establish clear rules for data and accountability.

Decide which information staff may enter into AI tools, which applications are approved and who is responsible for checking consequential outputs. High-impact decisions should have suitable human oversight, with additional safeguards where legal, financial, employment or customer consequences are involved.

5. Review roles and performance as the technology develops.

Use pilot results to determine whether responsibilities should change, whether employees need additional training and whether the business case justifies wider deployment. Where automation reduces demand for certain tasks, consider retraining or redeployment where feasible and handle workforce changes responsibly.

What African businesses need to consider

For businesses in African markets, the opportunity is not limited to copying the AI strategies of large global companies. Smaller firms may be able to use selected tools to improve customer response times, reduce administrative burdens or help employees perform research and routine tasks more efficiently.

However, the benefits are not automatic. Businesses need to consider the cost of reliable tools, connectivity, data quality, local-language performance, employee capabilities and the confidentiality of customer and company information.

The IMF has warned that countries with weaker digital infrastructure and skills may be less prepared to capture AI’s benefits. Its research also raises the possibility that uneven adoption could widen differences between countries and between workers who can benefit from AI and those who are more exposed to displacement.

For African businesses, this makes investment in people and digital capability particularly important. Companies that build skills and select suitable use cases may be better positioned to benefit than those that either reject the technology entirely or adopt it without a plan.

The productivity question matters as much as the jobs question

The debate about AI often centres on whether the technology will replace workers. For business leaders, that is only part of the calculation.

The more useful questions are whether AI can improve output quality, reduce avoidable costs, make employees more effective or allow the company to offer better services. Employers also need to consider who benefits from those improvements and how the organisation will support people whose work changes.

AI adoption should therefore be treated as an operating and workforce strategy, not simply a software purchase or a headcount exercise.

The business takeaway

The IMF’s estimate that almost 40% of global employment is exposed to AI signals a potentially significant transformation in work. It does not mean that 40% of jobs are certain to disappear.

Businesses that prepare carefully can assess where AI adds value, train their teams, protect sensitive information and make better-informed decisions about automation. Those that rely on the headline alone risk either overreacting to the threat or overlooking the opportunity.

The competitive advantage will not necessarily belong to the businesses that automate the most. It may belong to those that identify the right tasks, use AI responsibly and equip their people to work effectively as the technology changes.