Promise, Skepticism, and Their Implications for Southeast Asia

This dichotomy – significant potential coupled with limited deployment – is the focus of Capgemini’s latest research. The report, based on an April 2025 survey of 1,500 executives at large organizations across 14 countries, including Singapore, underscores the vital roles of trust and oversight in realizing value. Nearly three-quarters of executives surveyed believe the advantages of human involvement in AI workflows offset the associated costs. Furthermore, 90% perceive oversight as either beneficial or at least cost-neutral.

The takeaway is clear: AI agents are most effective when working with humans, rather than operating autonomously.

Early Stage Adoption

Approximately one-quarter of the surveyed organizations have launched agentic AI pilot programs, while only 14% have progressed to implementation. For the majority, deployment is still in the planning stages. The report highlights this growing divide between intention and preparedness, citing it as a significant impediment to capturing economic value.

While still emerging, real-world applications are beginning to surface. One example is a personalized shopping assistant capable of searching for items based on specific requests, generating product descriptions, answering inquiries, and adding items to a cart via voice or text commands. Although security concerns often prevent these tools from completing financial transactions, they already mirror many of the functions of a human assistant.

This development raises crucial questions about the future of traditional e-commerce websites. If AI can effectively manage tasks such as searching, comparing, and preparing purchases, will consumers still need to directly navigate online stores? For those who find complex websites overwhelming or difficult to use, an AI-driven interface could offer a simpler, more accessible alternative.

Defining Agentic AI

To clarify the hype surrounding agentic AI, *AI News* spoke with Jason Hardy, Chief Technology Officer for Artificial Intelligence at Hitachi Vantara, for insights on how Asia-Pacific enterprises should approach this technology.

Hardy defines agentic AI as “software that can decide, act, and refine its strategy on its own.” He elaborates, “Think of it as a team of domain experts that can learn from experience, coordinate tasks, and operate in real time. Generative AI creates content and is usually reactive to prompts. Agentic AI may use GenAI inside it, but its job is to pursue objectives and take action in dynamic environments.”

This distinction – between generating outputs and driving outcomes – is crucial for understanding the practical applications of agentic AI in enterprise IT.

The Drivers Behind Acceleration

Hardy attributes the growing interest in agentic AI to increasing scale and complexity. “Enterprises are drowning in complexity, risk, and scale. Agentic AI is catching on because it does more than analyse. It optimises storage and capacity on the fly, automates governance and compliance, anticipates failures before they occur, and responds to security threats in real time. That shift from ‘insight’ to ‘autonomous action’ is why adoption is accelerating,” he explains.

Capgemini’s research corroborates this view, finding that while confidence in agentic AI varies, early deployments are demonstrating value by automating routine but critical IT tasks.

Emerging Value Propositions

Hardy cites IT operations currently as the most compelling use case. “Automated data classification, proactive storage optimisation, and compliance reporting save teams hours each day, while predictive maintenance and real-time cybersecurity responses reduce downtime and risk,” he states.

The benefits extend beyond mere efficiency. These capabilities enable systems to preemptively detect problems, allocate resources more effectively, and contain security incidents more quickly. “Early adopters are already using agentic AI to remediate incidents proactively before they escalate, strengthening reliability and performance in hybrid environments,” adds Hardy.

For now, IT presents the most practical entry point, as its deployment yields measurable results and is central to how enterprises manage costs and risk.

Southeast Asia’s Starting Point

For organizations in Southeast Asia, Hardy emphasizes the importance of data quality. “Agentic AI delivers value only when enterprise data is properly classified, secured, and governed,” he explains.

Infrastructure is also a key factor, necessitating systems capable of supporting multi-agent orchestration, persistent memory, and dynamic resource allocation. Without this foundational layer, adoption will remain limited in scope.

Many enterprises may find it advantageous to begin with IT operations, where agentic AI can prevent outages and optimize performance, before expanding to broader business functions.

Reshaping Core Workflows

Hardy anticipates agentic AI will reshape workflows across IT, supply chain management, and customer service. “In IT operations, agentic AI can anticipate capacity needs, rebalance workloads, and reallocate resources in real time. It can also automate predictive maintenance, preventing hardware failures before they occur,” he notes.

Cybersecurity also holds immense promise. “In cybersecurity, agentic AI is able to detect anomalies, isolate affected systems, and trigger immutable backups in seconds, reducing response times and mitigating potential damage,” says Hardy.

These capabilities are not merely theoretical. Early deployments are already demonstrating how agentic AI can enhance reliability and resilience in hybrid environments.

The Need for New Skills and Leadership

Successful adoption hinges on developing new human skills. “Agentic AI will shift the human role from execution to oversight and orchestration,” Hardy asserts. Leaders will need to establish boundaries and monitor autonomous systems, ensuring they operate within ethical and organizational constraints.

For managers, this transition means less focus on administrative tasks and more emphasis on mentoring, innovation, and strategy. HR teams will need to cultivate governance skills such as auditing readiness and design new frameworks for effectively integrating agentic AI.

The impact on the workforce will be uneven. The World Economic Forum projects that AI could create 11 million jobs in Southeast Asia by 2030 while displacing nine million. Women and Gen Z are expected to experience the most significant disruptions, with over 70% of women and up to 76% of younger workers in roles vulnerable to AI.

This underscores the urgency of reskilling initiatives, with major investments already underway, including Microsoft’s $1.7 billion commitment in Indonesia and training programs across Malaysia and the wider region. Hardy stresses that capacity building must be inclusive, rapid, and strategic.

Looking Ahead

Looking three years into the future, Hardy believes many leaders will underestimate the speed of change. “The first wave of benefits is already visible in IT operations: agentic AI is automating tasks like data classification, storage optimisation, predictive maintenance, and cybersecurity response, freeing teams to focus on higher-level strategic work,” he says.

However, the larger surprise may lie in its impact at the economic and business model level. IDC projects AI and generative AI could add approximately $120 billion to the GDP of the ASEAN-6 by 2027. Hardy believes the implications are more far-reaching and will unfold more rapidly than many anticipate. “This suggests the impact will be much faster and more material than many leaders currently anticipate,” he concludes.

In Indonesia, over 57% of job roles are expected to be augmented or disrupted by AI, highlighting that transformation will not be confined to IT. It will fundamentally reshape how businesses are structured, how they manage risk, and how they create value.

Balancing Autonomy with Oversight

The Capgemini findings and Hardy’s insights converge on a central theme: agentic AI holds tremendous promise, but its practical success depends on striking a balance between autonomy and trust, and the continued necessity of human oversight.

The technology has the potential to help enterprises reduce costs, improve reliability, and unlock new revenue streams. However, neglecting governance, reskilling, and infrastructure readiness could jeopardize widespread adoption.

For Southeast Asia, the key question is not *if* agentic AI will take hold, but *how quickly* – and whether enterprises can effectively balance autonomy with accountability as machines assume greater responsibility for business decisions.