Artificial Intelligence is splitting employees into a minorities and majorities, a new work place survey shows.
A smaller group is using it to boost productivity while a larger group risk being left behind as technology changes the nature of jobs.
This is the central warning in PwC’s 2026 Global Workforce Hopes and Fears Survey, which paints a striking picture of a global labour market being reshaped not simply by AI adoption, but by unequal access to the opportunities it creates.
The survey, based on responses from 49,364 workers in 48 countries and 29 sectors in May and June, identifies a small group of “front-runners” accounting for 14 per cent of workers.
These employees combine strong AI capabilities with skills that are in high demand.
At the opposite end is the “engine room”, accounting for 56 per cent of the workforce.
These workers remain central to keeping organisations running, but have less access to AI, learning opportunities and career progression.
The divide is particularly significant because AI is moving rapidly from experimentation into everyday work.
According to the report, 64 per cent of workers now use AI in their jobs, up 10 percentage points from last year.
Daily use of generative AI has also risen from 14 per cent to 22 per cent.
The gains are uneven. More than half of the front-runner group use GenAI daily, compared with only 11 per cent of the core workforce.
Daily users are also more confident about their job security, with 68 per cent saying they feel very secure in their jobs.
The survey is particularly relevant to Kenya’s job market, characterised by a large informal economy and a persistent mismatch between skills available and opportunities.
The 2026 Economic Survey by the Kenya National Bureau of Statistics shows that although the economy created 822,100 jobs during the year, 87.2 per cent of these were in the informal sector.
Modern-sector wage employment stood at 3.3 million.
The numbers underline a fundamental problem: Kenya is creating jobs, but much of the expansion is happening outside the structured employment environment where workers are more likely to receive formal training, career development and social protection.
That makes the AI skills divide potentially more consequential.
The World Bank has previously identified a mismatch between the skills young people in Kenya acquire and those demanded by the labour market.
The PwC report suggests that the mismatch could take a new form as AI becomes embedded in recruitment, customer service, finance, communications, programming, marketing and other professional functions.
“The issue is not necessarily that machines will simply eliminate jobs. Rather, workers who know how to use AI alongside their existing expertise could increasingly become more valuable than those performing the same tasks without the technology,’’ the report reads.
That is already reflected in the global wage market. PwC says its 2026 AI Jobs Barometer found that workers with AI skills command a 62 per cent wage premium, up from 57 per cent in 2025.
“For Kenya, where wage employment remains a relatively small part of the workforce, that premium could deepen existing inequalities unless access to skills follows the technology.”
There are, however, signs that East Africa is not starting from a position of technological disadvantage.
PwC’s regional analysis found that 72 per cent of East African workers had used AI at work during the previous 12 months, compared with 54 per cent globally in the earlier survey.
Daily GenAI use stood at 38 per cent among the regional respondents. Workers also reported strong gains in productivity, quality and creativity from AI.
But exposure does not automatically translate into employability.
The regional study found that 71 per cent of workers reported access to learning and development resources, while 77 per cent said they had acquired new skills during the previous year.
Yet financial pressure remains acute: 77 per cent of East African respondents reported experiencing financial pressure.
The broader global findings point to the same tension. Only 51 per cent of workers said they had access to the learning and development resources they needed, down from 59 per cent a year earlier.
Among the 56 per cent “engine room” workforce, fewer than two in five reported having such access.
At the same time, only 34 per cent of workers globally said they could pay their bills and still have money left at the end of the month, down eight percentage points year-on-year.
Kenya’s earnings data provide another dimension to the challenge. KNBS estimates that real average earnings per employee rose to Sh678,800 in 2025 from Sh665,400 in 2024, with private-sector real earnings rising 3.9 per cent to Sh716,100.
Yet the growth in real earnings does not erase the structural vulnerability created by informal employment and limited access to quality jobs.
The International Labour Organisation has warned that Kenya’s AI transition must therefore be accompanied by investment in digital skills, practical training and access to opportunities beyond Nairobi and other technology centres.
It argues that generative AI is more likely to transform many jobs than simply eliminate them; with repetitive tasks declining while demand grows for digital fluency, analytical thinking and ethical awareness.
According to the survey, this shifts the responsibility beyond individual workers.
“Organisations need to redesign jobs and workflows, train employees to use the technology and create incentives for experimentation.”
