Automation strategy is often built around the formal workplace: defined roles, payroll records, structured training and measurable productivity. That frame misses a large part of the global labour market. Informal workers can face very different technology risks, with fewer protections and less access to the tools that would help them adapt.
One automation story does not fit every worker
Traditional automation and artificial intelligence do not affect work in the same way. Industrial automation is more directly associated with routine manual tasks, while generative AI is better suited to information processing and codifiable cognitive work. The result is a split exposure pattern: informal workers are often more vulnerable to traditional automation, while formal workers may have greater exposure to AI-driven task change.
This distinction matters for leaders operating across developing markets. A policy designed only for office-based AI adoption can overlook workers in agriculture, construction, transport, retail and small-scale services. Conversely, a programme focused only on industrial displacement may miss the rapid redesign of professional and administrative jobs.
Occupation matters more than employment label
The risk is not determined simply by whether someone works formally or informally. Occupational structure shapes the tasks performed and the technologies that can replace, reorganise or complement them. Two workers in the same sector may therefore experience automation very differently depending on their role, access to equipment and ability to learn new tasks.
Age, location and income can intensify the gap. Younger and older informal workers in rural areas, especially those with low or middle earnings, may have fewer pathways into retraining. The commercial implication is clear: companies cannot assume that a general digital-skills programme will reach the people most exposed to operational change.
Adoption plans need a worker transition map
Before deploying automation, businesses should map the workforce beyond direct employees. That includes contractors, distributors, platform workers, small suppliers and other people whose livelihoods depend on the operating model. The map should identify which tasks may disappear, which will change and where new work could emerge.
Three questions make the exercise practical:
- Which tasks are technically replaceable, and which still depend on local judgement or physical adaptability?
- Who can access training, devices, connectivity and recognised credentials?
- What transition support is available to workers outside the formal payroll?
This approach turns social risk into an operational design issue. If a supplier network loses capability faster than it gains new skills, automation may reduce resilience even when unit costs fall.
Training must be reachable, not merely available
Formal employees are more likely to receive employer-funded learning and protected time to use it. Informal workers may face immediate income loss when attending training, and many programmes require digital access or qualifications they do not possess. Effective support therefore needs flexible delivery, short modules, portable credentials and, where possible, income protection during transition.
Partnerships with local institutions, cooperatives and industry groups can help businesses reach workers who sit outside conventional HR systems. Procurement standards can also reward suppliers that document retraining and safe technology adoption instead of treating labour adjustment as an external cost.
Exposure is not a forecast of job loss
Measures of automation risk usually estimate whether tasks are technically replaceable. They do not predict the timing or scale of actual job losses. Adoption costs, regulation, infrastructure, customer preferences and business models all affect what companies implement. New tasks and occupations may also emerge as old ones change.
Leaders should therefore avoid presenting exposure scores as inevitable outcomes. The useful value of the evidence is diagnostic: it shows where transition capacity is weakest and where a uniform automation policy could deepen existing inequality.
A better governance model
Boards can improve oversight by separating formal and informal workforce indicators while keeping them in the same transformation dashboard. Productivity gains should be reviewed alongside supplier resilience, training access, income disruption and pathways into new roles.
The strategic goal is not to slow every form of automation. It is to ensure that technology adoption strengthens the operating system around the business rather than shifting hidden costs onto workers with the least protection. A two-track policy—one for formal employment and one for informal work—offers a more realistic starting point.