IBM explores the potential of agentic AI

A recent ASEAN Agentic AI Virtual Roundtable organised by IBM has explored the potential of agentic AI to shape future-ready industries. The discussion centred on how this technology, which is capable of autonomous decision-making and goal-driven actions, is enhancing business productivity, improving operational efficiency, and reducing costs.

Catherine Lian, GM & Technology Leader, IBM ASEAN, observed that there is a "surge of interest in autonomous agents, ushering in the future of work," while Gartner has predicted that by 2028, 33% of enterprise software applications will include agentic AI, up from less than 1% in 2024, enabling 15% of day-to-day work decisions to be made autonomously.

"At IBM, we believe this new way of working will help unlock a new economic potential for generative AI," Lian said, sharing that across ASEAN, AI agents represent a pivotal advancement and offer transformative potential, but also bring challenges that require careful consideration.

The roundtable highlighted how agentic AI tools can help businesses remain competitive in the era of intelligent automation. McKinsey has estimated that agentic AI could generate up to US$4.4 T of economic value annually across industries.

Source: IBM CEO Study 2025. Global
findings.

Anup Kumar, Distinguished Engineer and Head of Client Engineering, IBM Asia Pacific, shared some insights from the IBM Institute for Business Value’s 2025 CEO Study. He noted that the biggest challenge that ASEAN CEOs face is "putting the silos together", especially when things are moving very fast.

The research revealed that CEOs surveyed globally expect the growth rate of AI investments to more than double in the next two years, and 61% confirm they are actively adopting AI agents today and preparing to implement them at scale.

Further, by 2027, 85% of surveyed CEOs expect their investments in scaled AI efficiency and cost savings to have returned a positive ROI, while 77% expect to see a positive return from their investments in scaled AI growth and expansion.

ASEAN CEO perspectives, on the other hand, included:

- Nearly six in 10 (57%) ASEAN CEOs say they are actively adopting AI agents and prepared to scale them across their organisation (52% in Singapore; 61% globally)

- Six in 10 ASEAN CEOs say they have prepared their employees for the cultural and operational shifts from AI agents

- Seven in 10 (71%) of ASEAN CEOs will prioritise AI use cases based on return on investment (ROI) 

- Over two thirds (61%) of ASEAN CEOs said AI is changing aspects of the business that they consider core

- Slightly over half (51%) of ASEAN CEOs said they have disconnected technology pieces across the organisation as a result of the pace of change

- Top three priorities for ASEAN CEOs in the next three years: Forecast accuracy (44%); product & service innovation (40%); ecosystems & partnership (39%).

- Barriers to innovation: Lack of expertise (44%); risk aversion (44%); inadequate technology (42%).

Indonesia 

- Over six in 10 (63%) of CEOs surveyed cited that competitive advantage depends on who has the most advanced generative AI. 

- About six in 10 (61%) CEOs surveyed cited that they are actively adopting AI agents and prepared to scale them across their organisation. 

- Top three barriers to innovation: Lack of expertise (57%); risk aversion (44%); inadequate technology (50%). 

Malaysia 

- Six in 10 CEOs surveyed cited that competitive advantage depends on who has the most advanced generative AI. 

- Nearly seven in 10 (69%) of CEOs surveyed cited that they are actively adopting AI agents and prepared to scale them across their organisation. 

- Top barriers to innovation: Lack of expertise (35%); risk aversion (35%); inadequate technology (45%). 

Philippines 

- Fifty-five percent of CEOs surveyed cited that competitive advantage depends on who has the most advanced generative AI. 

- The same percentage of CEOs surveyed cited that they are actively adopting AI agents and prepared to scale them across their organisation. 

- Top barriers to innovation: Lack of expertise (45%); risk aversion (48%); inadequate technology (40%). 

Singapore 

- Six in 10 CEOs surveyed cited that competitive advantage depends on who has the most advanced generative AI. 

- A little over half (52%) of CEOs surveyed cited that they are actively adopting AI agents and prepared to scale them across their organisation. 

- Top barriers to innovation: Lack of expertise (45%); risk aversion (45%); inadequate technology (37%). 

Kumar also described the AI customer journey, which typically starts with chatbots, moves on to AI assistants, and then to agents.  

Participants at the roundtable had much to say about agentic AI:

- Agentic AI is designed to augment existing infrastructure, not disrupt it. Agentic AI enhances current capabilities by introducing intelligent, autonomous decision-making within established systems. 

- Orchestration and automation are the secret weapons to unlock AI productivity — enabling businesses to connect and manage vast networks of agents and assistants across broad enterprise technology stacks. Lian said "agents must be able to work seamlessly across a vast web of applications, data and systems that underpin complex enterprise technology stacks". 

"This means that they can orchestrate, integrate, and automate," she said. "And these are really the key movers to really run the actions towards the ROI that they're seeking."

- IBM introduced a new orchestrator agent feature to IBM watsonx Orchestrate for employee needs that uses finetuned foundation models within an agentic architecture to provide advanced reasoning capabilities and autonomous decision-making. 

Dr Clifton Phua, Director of Labs, BizTech Group, Infocomm Media Development Authority (IMDA) said: “Singapore believes in harnessing AI for public good, to create positive impact across society and daily life. IMDA supports this vision by fostering innovation through responsible AI deployment. 

"For agentic AI, we are collaborating with industry partners to test use cases whilst emphasising safety by design. These use cases will help workers upskill, and become proficient in AI, while enhancing productivity for companies. When we have AI systems that can collaborate meaningfully, every business process becomes an opportunity for innovation –
much like upgrading from a map to a navigator, not just showing the path, but actively helping to find the best route forward.”

Henke Yunkins, Director of Regulation & Ethics, Indonesia Artificial Intelligence Society (IAIS) noted: "Building public trust in AI requires more than good intentions, it demands practical implementation of transparency principles. Most AI systems still function as opaque ‘black boxes,’ making it difficult for users, customers, and regulators to understand how decisions are made. 

"To move beyond theoretical transparency, organisations need to operationalise explainability: designing systems that can clearly communicate their decision logic, risks, and limitations in human-understandable terms. Trust isn’t built through compliance checklists— it’s earned when people can see, question, and understand how AI affects them. That’s the foundation for ethical deployment, competitive advantage, and sustainable innovation in agentic AI.”

Said Jack Madrid, President and CEO, IT and Business Process Association of the Philippines (IBPAP): “With 10% of IT-BPM organisations having fully implemented generative AI and another 56% actively integrating it into their operations, our industry is already experiencing rapid digital transformation. We expect this to continue with the emergence of agentic AI as more organisations aim to unlock even greater value through orchestrating complex processes, optimising workflows, and driving intelligent automation.

"At the same time, the widespread adoption of agentic AI will shift workforce demand from repetitive, transactional roles to higher-value, AI-enhanced functions, which highlights the urgent need for large-scale workforce upskilling to keep our sector globally competitive. The Philippines is in a sweet spot to scale agentic AI—not just to automate, but to elevate how we deliver high-volume, knowledge-driven services.”

BPM stands for business process management.

The roundtable concluded that upskilling organisations can build public trust in AI through practical solutions.