The global job market is experiencing a fundamental split driven by how organisations deploy artificial intelligence, according to a comprehensive PricewaterhouseCoopers study released this week. Rather than simply automating work away, AI is rewarding companies that use the technology to amplify human capabilities while penalising those that view it merely as a cost-reduction tool. This divergence suggests that the future workplace will belong to organisations that strategically integrate AI with human expertise rather than those seeking to replace workers entirely.
The evidence is striking. Roles demanding specific AI competencies—such as machine learning specialists and prompt engineers—expanded at nearly eight times the pace of the overall job market in 2025, growing 69% compared to just 9% for employment broadly. These positions command substantial wage premiums too, with compensation packages now running 62% above comparable non-AI roles, up from 57% the previous year. The gains are not uniform across sectors, however. Consumer-facing industries offer premium salaries for AI specialists reaching 118% above baseline, while government and public sector organisations lag significantly at just 16% above standard rates.
What distinguishes the most successful companies is their approach to technology integration. PwC's research, drawing on over one billion job postings across 27 countries and territories, found that firms most heavily exposed to AI employment actually increased their headcount by 52% between 2018 and 2025. In contrast, companies with minimal AI adoption grew employment by only 36%. This counterintuitive finding demolishes the narrative that artificial intelligence inevitably destroys jobs. Instead, the data suggests that strategic AI implementation opens new business opportunities and market segments that require expanded workforces.
The nature of work itself is undergoing profound transformation, particularly at entry level. Traditionally, junior positions served as apprenticeships where new workers learned through handling routine, repetitive tasks. That model is now antiquated. Entry-level roles increasingly demand competencies once reserved for senior professionals—creativity, judgment, ethical reasoning, empathy, and leadership. Positions requiring these distinctly human skills have grown 35% since 2019, while conventional entry-level roles without such requirements have contracted by 10%. This shift has troubling implications for career progression and talent development across industries.
Corporate leaders are acutely aware of this transformation. Nearly half of chief executives surveyed by PwC expect AI adoption to reduce hiring of junior staff over the next three years, compared to just 12% who anticipate similar cuts at senior levels. This represents a dangerous inversion of traditional organisational hierarchies. As Pete Brown, PwC's global workforce leader, notes, AI is simultaneously removing the routine work that historically trained new talent while dramatically increasing demand for advanced judgment and adaptability much earlier in careers. Organisations have not yet adapted their talent development strategies to this reality, creating potential skills shortages and career pathway challenges.
Certain professional categories illustrate the profound productivity gains possible when AI augments rather than replaces human expertise. Radiologists and recruiters—roles where AI enhances rather than eliminates human judgment—saw employment growth twice as fast as positions like IT service managers or medical secretaries, where AI primarily makes tasks simpler for less-skilled workers to perform. Financial analysts provide a particularly instructive case study. Rather than facing displacement, these professionals gained access to powerful new analytical tools enabling far more sophisticated analysis. Employment in this field has continued climbing as new specialisations emerge commanding premium compensation. The pattern is unmistakable: jobs where AI handles routine elements while humans focus on judgment, strategy, and interpretation thrive.
Sectoral patterns reveal significant variation in AI adoption trajectories. Technology, media and telecommunications sectors led AI-driven job growth last year at 11%, followed by professional services at 6%. Healthcare, despite its enormous potential for AI applications, lagged dramatically at under 1% growth. This disparity suggests either slower AI implementation in healthcare or different employment dynamics in that sector. The geographic and sectoral imbalances hint at future competitiveness challenges, as regions and industries moving faster on AI gain productivity and innovation advantages.
Productivity gains correlate strongly with AI exposure levels. Companies most heavily invested in AI achieved labour productivity growth of 34% between 2018 and 2025, substantially outpacing the 24% gains recorded by firms with minimal AI engagement. The top quintile of AI-exposed companies achieved truly exceptional results: 163% productivity improvements relative to 2018 levels, nearly five times the average across all AI-exposed firms. These numbers reflect not simply doing the same work faster, but fundamental changes in capability, output quality, and business model innovation enabled by human-AI collaboration.
The wage premium gap reflects market recognition of this value creation disparity. Organisations that successfully leverage AI to enhance human expertise can afford to compensate employees substantially higher, as the productivity and innovation gains justify elevated salary levels. Conversely, companies pursuing automation-first strategies face wage pressure and competitive disadvantages. This creates a self-reinforcing cycle where talent gravitates toward organisations implementing AI thoughtfully, leaving cost-focused competitors struggling to attract skilled workers and falling further behind on productivity metrics.
For Malaysian and Southeast Asian organisations, these findings carry particular significance. Many regional firms occupy intermediate positions in global supply chains and service delivery networks, making them vulnerable to displacement by AI-augmented competitors elsewhere. However, the PwC data suggests opportunity for companies that proactively rethink their talent strategies. Rather than viewing AI primarily as a tool for reducing labour costs—the traditional approach in developing economies—forward-thinking regional organisations can position themselves as centres of human expertise amplified by artificial intelligence. This requires investment in workforce upskilling, particularly in judgment-based competencies and creative problem-solving that complement AI capabilities.
The implications for Southeast Asian talent development and education systems are profound. Current vocational training and university curricula often emphasise technical skills and routine competencies. To compete effectively, educational institutions must pivot toward developing judgment, ethical reasoning, creative thinking, and adaptive problem-solving from entry level onwards. The old model of apprenticeship through routine work no longer applies. Instead, early-career professionals must possess advanced cognitive and interpersonal capabilities from day one. This represents both challenge and opportunity for the region's workforce development strategies.
Looking forward, organisations globally must fundamentally reconsider their AI strategy and talent philosophy. The companies pulling furthest ahead on productivity and growth are those using artificial intelligence to amplify human expertise, accelerate innovation, and create entirely new value sources rather than simply reducing headcount. Joe Atkinson, PwC's global chief AI officer, emphasises that winning in the AI age is not merely about technology deployment but about human skills. The more extensively AI is deployed, the more distinctly human expertise becomes valued. This insight should reshape how regional and global organisations approach workforce investment, talent acquisition, and business strategy in coming years.
