Artificial intelligence is entering a pivotal new phase as enterprises across Asia move from experimentation to widespread deployment. Early interest centred on generative models that produced text or accelerated routine tasks. Today, organisations are exploring systems that can reason, plan, and execute with far less oversight. This shift reflects growing confidence in AI’s ability to support operational resilience, strengthen customer experience, and enhance real-time decision-making in increasingly complex environments.
The evolution is happening as companies across the region face rising pressure to increase efficiency, manage costs, and respond to rapid economic and technological change. Many are reassessing established ways of working and questioning whether traditional automation remains sufficient. Agentic AI, which enables autonomous digital systems to execute multi-step workflows, is emerging as a practical solution to these challenges.
Industry research points to accelerating momentum. IDC forecasts that AI investment in Asia-Pacific excluding Japan will exceed US$78 billion by 2027, mainly driven by demand for autonomous decision-making and AI-enabled business applications. The availability of more powerful models reinforces this growth, rapid improvements in orchestration frameworks, and a shift in leadership mindset as AI becomes a core business capability rather than a standalone experiment.
Regional leaders are already signalling this shift. Jaime Vallés, Vice President for Asia Pacific and Japan at Amazon Web Services, has observed that enterprises are moving beyond trials and scaling systems to improve efficiency, strengthen resilience, and deliver better outcomes for customers. His perspective mirrors trends across the region as organisations prepare for a period of significant transformation.
Understanding the shift from generative AI to autonomous agents
The transition from generative AI to agentic AI represents a fundamental change in how organisations view automation. Generative AI delivered clear productivity benefits in content creation, analysis, software development, and customer interaction. However, these models often required human review to ensure accuracy and compliance. Agentic AI systems take this further by evaluating objectives, formulating plans, and completing tasks across integrated applications.
This makes them suitable for more complex and high-volume business processes. Industries such as financial services, telecommunications, logistics, and retail are exploring how autonomous agents can manage repetitive or multi-step tasks that previously relied on specialist teams. Their ability to combine reasoning with execution increases their usefulness in real operational settings.
Examples across Asia demonstrate the speed of this shift. In Australia, Lendi Group has adopted agents to manage documentation and administrative processes, saving thousands of hours of manual work each year. LG Electronics in Korea has reported a substantial acceleration in market analysis workflows, reducing tasks that previously took days to under an hour. These early deployments demonstrate how autonomous systems can deliver measurable results when applied to well-defined, data-rich processes.
Research firms expect this trend to continue. Gartner predicts that autonomous agents will be embedded in the operational workflows of more than 20% of large enterprises by 2026. This growth is driven by demand for consistent decision-making, streamlined processes, and faster response times. For leaders in Asia, the priority is understanding where autonomy strengthens operations, where human oversight remains necessary, and how to ensure that systems behave reliably at scale.
Why agentic AI is reshaping enterprise productivity
Productivity is a central reason driving enterprise interest in agentic AI. Many organisations across Asia operate in competitive markets characterised by rising labour costs, talent shortages, and heightened expectations for real-time service. Generative AI delivered improvements in knowledge work, but agentic AI addresses deeper structural inefficiencies by handling tasks end-to-end.
Autonomous agents can analyse information, apply predefined rules, make decisions, and execute actions across systems without human intervention. This reduces bottlenecks and shortens cycle times, particularly in processes involving extensive documentation, multi-system workflows, or strict compliance requirements. Teams can then redirect their efforts towards more strategic responsibilities, such as customer engagement, problem-solving, and innovation.
Industry studies reflect this advantage. McKinsey’s research on digital workers found that organisations deploying autonomous systems achieved productivity improvements of up to 50% in targeted processes. These gains were most pronounced in operations, finance, and service management—areas where repetitive tasks are common, and accuracy is essential. This aligns with developments in Asia, where companies are scaling automation beyond pilot projects and integrating agents into broader transformation strategies.
Vallés has observed this shift in mindset across Asia-Pacific, noting, “Instead of chasing AI for basic productivity wins, they’re putting it right into their strategy and customer experience because they know that’s where the real leverage is, using it to scale faster and shape how they serve customers.”
Consistency is another important consideration. Autonomous agents apply rules uniformly, reducing human variability and strengthening compliance. This is particularly valuable in sectors such as healthcare, insurance, and banking, where documentation errors or inconsistent decisions can create significant risks. By enforcing repeatable, traceable actions, agentic systems help organisations maintain quality while managing rising work volumes.
The introduction of autonomy also raises questions about measurement and accountability. Leaders must define how productivity should be calculated in a hybrid environment where humans and agents collaborate. They also need to determine where oversight is required and how job roles will evolve. These questions highlight the need for clear governance frameworks that balance innovation with risk management.
How leadership is evolving in an era of autonomous systems
The rise of agentic AI has brought leadership into sharper focus. Successful deployment depends not only on technical capability but also on team readiness, alignment of organisational culture, and clarity of governance. Leaders are now expected to guide their organisations through a period of change that affects processes, roles, and expectations.
Digital literacy is becoming a core requirement. Microsoft’s 2024 Work Trend Index notes that leaders in Asia are increasing investments in AI training, recognising that employees need to understand how to work effectively with autonomous systems. This involves not only technical skills but also the ability to supervise, validate, and collaborate with AI-driven processes. Governments across Singapore, South Korea, India, and Australia are supporting this shift through national AI and digital skills programmes.
Governance is an equally important area. As autonomous systems take on greater responsibility, organisations must define clear rules for decision-making, data use, and oversight. This includes ensuring transparency in how algorithms operate, establishing escalation pathways for exceptions, and maintaining audit trails. Enterprises with strong governance frameworks tend to adopt AI more confidently and sustainably, as they can manage risks without slowing innovation.
Organisational culture also influences the success of AI adoption. Companies that encourage experimentation, iterative learning, and cross-functional collaboration tend to integrate new technologies more effectively. In contrast, rigid structures and siloed processes can hinder progress. Leaders who promote open communication and align teams around shared goals create an environment where AI can drive meaningful impact.
As mentioned by Vallés, preparing organisations to use AI responsibly is becoming a defining skill for modern leaders. “Leadership today is measured by more than how advanced the technology is. What matters just as much is the ability to get teams, policymakers, and boards ready to innovate responsibly with AI, giving them the knowledge, confidence, and practical tools to make decisions that enable tangible progress.”
These shifts illustrate how leadership responsibilities are expanding. Autonomous systems require leaders to think beyond short-term deployment and consider long-term workforce strategy, capability building, and organisational resilience. Vallés has noted that leadership maturity must evolve alongside technological advancement, particularly in areas such as digital fluency and responsible governance. His observations reflect the broader challenges and opportunities facing organisations as they navigate this transition.
Why infrastructure decisions now shape long-term competitiveness
Effective use of agentic AI relies heavily on a strong technical foundation. Autonomous systems require scalable computing power, consistent access to high-quality data, and seamless integration with enterprise applications. As a result, infrastructure decisions made today will significantly influence organisational competitiveness over the coming decade.
Cloud platforms are central to this shift. AI workloads require elastic capacity and advanced computing resources that are difficult to maintain on traditional on-premises systems. Accenture research shows that 87% of large enterprises in the Asia-Pacific plan to increase cloud spending specifically to support AI. This reflects growing awareness that legacy environments cannot meet the performance and scalability requirements of agentic AI.
Data architecture is another critical component. Autonomous systems depend on reliable, well-governed, and accessible data. Fragmented datasets or inconsistent data standards undermine performance and limit the effectiveness of AI-driven decision-making. Many organisations are now prioritising unified data platforms, improved quality controls, and governance models that support secure sharing across business units and markets.
Regional regulatory considerations also influence infrastructure choices. Asia’s regulatory landscape varies widely, with differing rules on data localisation, privacy, and cross-border transfers. Enterprises operating across multiple APEC markets must design flexible architectures that meet local compliance requirements without compromising efficiency. This often requires a hybrid approach that balances local control with centralised capabilities.
The long-term implications are becoming clearer. Organisations that invest early in scalable, compliant, and AI-ready infrastructure will be well-positioned to integrate future capabilities and remain competitive. Those who delay may face rising technical debt, higher costs, and slower innovation cycles. The connection between infrastructure and strategic agility is strengthening as AI adoption accelerates.
As shared by Vallés, the decisions enterprises make about their cloud and data foundations today will shape who moves ahead in the region. “The choices businesses make about infrastructure now, particularly about secure, scalable cloud, will determine who captures the next wave of AI-driven growth in Asia-Pacific.”
Collaboration and ecosystem growth in the AI economy
The expansion of agentic AI has highlighted the importance of collaboration across the technology ecosystem. Enterprises, startups, research institutions, and industry partners each play a role in shaping how AI evolves and how quickly organisations can deploy new capabilities.
Startups are a key source of innovation, particularly in areas such as agent orchestration, vertical-specific models, and workflow automation tools. Research from Google and Temasek shows that Southeast Asia’s digital economy is projected to reach US$300 billion by 2025, with AI-driven startups contributing significantly to enterprise transformation. Their agility allows companies to experiment more quickly and access capabilities that complement their internal systems.
Large organisations rely on partnerships with cloud providers, software vendors, and consultancies to support implementation at scale. These collaborations are essential in sectors with complex regulatory or operational requirements, such as finance, manufacturing, and healthcare. Research institutions across the region—including NUS, KAIST, and the University of Melbourne—are also expanding programmes focused on autonomous systems, robotics, and human-AI interaction. These initiatives help strengthen the talent pipeline and support long-term innovation.
Governments in the region are encouraging a collaborative approach through national AI strategies and investment programmes. Singapore’s National AI Strategy, Japan’s AI governance frameworks, and Australia’s digital capability initiatives are creating environments where businesses can test, refine, and scale AI safely. These policies often include incentives for experimentation, regulatory sandboxes, and support for upskilling.
Ecosystem collaboration is becoming a defining feature of Asia’s AI progress. Vallés has highlighted the region’s openness to experimentation and its diversity of markets as strengths that enable companies to adopt new technologies quickly. His observations reflect broader trends showing that shared expertise and cross-sector partnerships can achieve outcomes that individual organisations might struggle to deliver on their own.
