Finding ways to adopt artificial intelligence (AI) has become a priority for many businesses, with companies across industries rushing to harness its promise of higher productivity, lower costs and new growth opportunities.
However, AI’s rising energy consumption and growing public scrutiny over workforce disruption are adding to governance concerns, now emerging as one of the most significant challenges facing corporate leaders.
“The challenge when it comes to AI governance is that the societal risks it brings cut across all geographies, sectors and risks. This is compounded by nearly all businesses playing a role in AI’s societal impact,” said Steven Okun, founder and chief executive officer of APAC Advisors, a Singapore-based strategic advisory firm focused on sustainability, public policy and stakeholder engagement.
While organisational leaders are focused on the productivity and efficiency gains that AI can deliver, Okun said these benefits come with trade-offs that are often overlooked.
Workforce displacement from rising AI technology remains one of its most pressing societal challenges. Contrary to the perception that automation primarily affects low-skilled jobs, AI is increasingly disrupting white-collar and knowledge-based work, he said, pointing to DBS Bank’s plans to cut around 4,000 jobs over three years while creating 1,000 new AI-related positions.
The Singapore bank announced in February 2025 that it plans to reduce around 4,000 temporary and contract roles over the next three years as AI increasingly takes on tasks traditionally performed by humans, while creating about 1,000 new AI-related positions.
Surveys show that Singaporeans are increasingly worried about workplace displacement. Accounting standards organisation ACCA’s recent Global Talent Trends Report found that 31 per cent of respondents in Singapore now cite “jobs being replaced by technology” as their biggest workplace fear, up from just 8 per cent in 2025, reflecting growing unease even as companies accelerate digital transformation.
Eight factors expected to drive career transformation in the accountancy profession over the next 10 years. Image: Career Paths Reimagined / ACCA
While new opportunities from AI will emerge, Okun noted that the benefits are likely to be concentrated in highly skilled, technology-focused roles, potentially leaving routine and entry-level workers behind.
Without real investment in reskilling from businesses themselves, the fallout extends far beyond lost jobs, deepening inequality, eroding social cohesion, and threatening the stability that underpins Singapore’s appeal as a business hub, he said.
Beyond employment concerns, AI also raises questions around fairness, bias and access. Women in Singapore, for example, are disproportionately represented in routine roles that are more vulnerable to automation, while men tend to dominate sectors expected to benefit most from AI adoption. Across Southeast Asia, around 70 per cent of female workers are employed in occupations with a high exposure to automation risks.
Okun also highlighted the often-overlooked labour underpinning many AI systems, including content moderators and data labellers in lower-income countries whose work remains largely invisible despite being critical to the development of AI models.
“AI is often framed as the great equaliser. The reality is the opposite,” he said. “Without deliberate action, AI’s gains flow to those already ahead, widening Singapore’s divides rather than closing them.”
The environmental dimension of the trilemma is equally significant. As AI adoption accelerates, demand for data centres continues to grow, increasing pressure on energy systems and raising concerns over land, water and electricity consumption. Together with workforce and inequality challenges, these factors are likely to shape future public and regulatory responses to AI deployment.
Reskilling needed
While Singapore has sought to balance innovation and responsible development through initiatives such as the Infocomm Media Development Authority’s Model AI Governance Framework and the establishment of a National AI Council chaired by Prime Minister Lawrence Wong, Okun noted that much of the current approach remains voluntary and focused on enabling innovation.
The country’s national lifelong learning initiative SkillsFuture is among several programmes providing avenues for individuals to reskill, but Okun argued the responsibility should not fall on workers alone.
“Singapore’s response will fall short unless businesses and investors step up too,” he said.
The challenges facing businesses are also becoming evident within professional services sectors, where AI is reshaping traditional career pathways and raising new governance concerns.
Dean Hezekiah, policy and insights manager at ACCA said the impact of AI on jobs is difficult to predict because the technology continues to evolve and affects different occupations in different ways. However, entry-level roles are expected to undergo some of the most significant changes, he said.
ACCA’s report Career Paths Reimagined: The Changing World of Work, published in January, found that many entry-level positions are shifting away from routine processing tasks towards higher-value activities such as data analysis and insight generation.
“This trend is expected to affect accountants working in both practice and business,” he said, adding that similar shifts are taking place across other industries.
Hezekiah noted that recruitment is one area that AI is increasingly being leveraged in. The firm’s Global Talent Trends Report 2026 found that employers are increasingly using AI to filter and shortlist candidates for entry-level positions, where application volumes are typically highest. More strategic and senior hiring decisions, however, continue to rely heavily on human judgement.
“
Without deliberate action, AI’s gains flow to those already ahead, widening Singapore’s divides rather than closing them.
Steve Okun, founder and chief executive officer, APAC Advisors
Beyond workforce impacts, Hezekiah identified data governance as one of the most significant AI-related risks facing the accountancy profession.
The growing use of AI-powered software solutions has added another layer of complexity to an already intricate data governance landscape. Many software-as-a-service providers now include terms allowing them to use AI to improve existing products or develop new services, raising questions about how client information is stored, processed and utilised.
As accounting firms increasingly rely on these platforms to serve clients, he said there is a risk that AI-driven data usage may not align with client expectations or regulatory requirements.
The issue also underscores the need for organisations to establish clear governance frameworks and ethical principles around AI adoption and ensure AI deployment is aligned with organisational values and broader societal expectations.
From adoption to resilience
As businesses move beyond experimentation into large-scale AI deployment, both experts say governance frameworks will need to evolve just as quickly as the technology itself.
For Okun, the starting point is recognising that AI risk is no longer hypothetical. Instead, it should be treated as a core business continuity issue, one that sits alongside climate and geopolitical risks in boardroom planning.
He argued that companies should begin stress-testing their strategies against a range of possible futures where access to AI is no longer guaranteed, whether due to rising energy costs, infrastructure constraints or regulatory intervention.
“AI access may not always be unlimited as risks mount. Companies need to stress-test their plans to ensure business continuity in a world disrupted by climate impacts, rising electricity prices and regulation where access to artificial intelligence is scarce or expensive,” he said.
Like climate risk, he added, AI risk requires forward-looking scenario planning that tests how resilient a business model is under different conditions of availability, cost and constraint.
Central to this approach is what he calls the “Responsible AI Trilemma”, the interconnected challenges between environmental impact, workforce disruption and inequality. These pressures, he said, are increasingly shaping public sentiment and regulatory direction, particularly around the energy-intensive infrastructure that powers AI systems.
Despite their significance, these trade-offs are rarely addressed at a strategic level, he said, adding that leadership teams should integrate the trilemma into long-term planning and assess how it could affect operations, profitability and licence to operate.
While corporate boards reassess risk frameworks, attention is also turning to how governments, employers and educators can cushion the workforce impacts of AI-driven change.
Despite rising anxiety around job displacement, Hezekiah said individuals can still meaningfully prepare for an AI-shaped future by focusing on skills that complement technology such as curiosity, storytelling, strategy and ethical judgement.
“These are also areas where integrating AI can provide significant uplift, so it is helpful to get hands-on with AI tools and build that confidence where possible,” he said.
However, he stressed that responsibility cannot sit with workers alone. Employers, educators and policymakers must work together to build an AI-ready workforce, not only through technical training programmes but also by strengthening critical thinking and lifelong learning capabilities.
Such skills, he said, are essential for ensuring individuals can effectively interpret AI outputs and make informed judgments rather than relying on automated systems blindly.
You are enjoying free access to this article, thanks to the support of our strategic partner, National Volunteer and Philanthropy Centre (NVPC) and the Company of Good.
Join the movement of leaders shaping impact. Find out more about what it takes to be nationally recognised as a Company of Good.
