The Rise of Agentic AI: Transforming Enterprises for a New Era

Agentic AI is being heralded as the next significant evolution in artificial intelligence, poised to redefine how enterprises operate. According to a report from the Capgemini Research Institute, Agentic AI could unlock an astounding US$450 billion in economic value by 2028. However, despite its potential, only 2% of organizations have scaled its use, and trust in AI agents is waning.

Understanding the Gap: High Potential vs. Low Deployment

This tension between the promising potential of Agentic AI and its limited adoption is the focus of Capgemini’s new research. Based on a survey conducted in April 2025 with 1,500 executives from large organizations across 14 countries, including Singapore, the report underscores the significance of trust and oversight in realizing the full value of AI technologies. Almost three-quarters of executives believe the advantages of human involvement in AI workflows outweigh the costs, while 90% view oversight as either beneficial or at least cost-neutral.

Early Steps and Slow Progress in Adoption

Currently, about 25% of organizations have initiated pilot programs for Agentic AI, but only 14% have moved into full-scale implementation. Most companies remain in the planning phase, indicating a widening gap between intent and readiness, which is a significant barrier to capturing economic value.

Real-World Applications: A Personal Shopping Assistant Example

While the technology may seem theoretical, real-world applications are beginning to take shape. For instance, a personal shopping assistant can search for items based on specific requests, generate product descriptions, answer queries, and compile shopping carts using voice or text commands. Although these tools often refrain from processing financial transactions for security reasons, they replicate many functions traditionally performed by human assistants.

Redefining the Role of Traditional Websites

This evolution prompts a critical question: If AI can execute tasks like searching, comparing, and preparing purchases, will consumers still find value in navigating online stores directly? For users overwhelmed by busy websites, AI-driven interfaces may present a simpler, more user-friendly alternative.

Defining Agentic AI: Insights from Industry Leaders

To clarify the concept, AI News consulted Jason Hardy, Chief Technology Officer for Artificial Intelligence at Hitachi Vantara. He defines Agentic AI as software capable of making decisions, taking action, and refining its strategy autonomously. “Think of it as a team of domain experts that can learn from experience, coordinate tasks, and operate in real time,” Hardy explained. This sets Agentic AI apart from generative AI, which primarily creates content in response to prompts.

Why Adoption is Accelerating: Scale and Complexity

According to Hardy, the acceleration in adoption is driven by the increasing scale and complexity of enterprise operations. “Organizations are grappling with complexity, risk, and scale. Agentic AI is gaining traction because it goes beyond mere analysis; it optimizes storage and capacity dynamically, automates governance and compliance, anticipates failures, and responds to security threats in real time,” he noted. Capgemini’s research corroborates this, revealing that while confidence in Agentic AI varies, early deployments are proving beneficial in handling routine yet essential IT tasks.

Identifying Emerging Value: The IT Operations Case

Hardy points to IT operations as the most compelling use case for Agentic AI. Automated data classification, proactive storage optimization, and compliance reporting can save teams significant time daily, while predictive maintenance and real-time cybersecurity responses mitigate downtime and risk. The ability to detect issues before they escalate enhances reliability and performance in hybrid environments.

Southeast Asia: Prioritizing Data Quality

For organizations in Southeast Asia, Hardy emphasizes that the first priority must be ensuring data quality. “Agentic AI delivers value only when enterprise data is properly classified, secured, and governed,” he stated. Infrastructure is crucial; Agentic AI requires systems capable of multi-agent orchestration, persistent memory, and dynamic resource allocation to succeed.

Reshaping Workflows Across Industries

Hardy anticipates that Agentic AI will reshape workflows in IT, supply chain management, and customer service. “In IT operations, Agentic AI can predict capacity needs, rebalance workloads, and reallocate resources in real time. It can also automate predictive maintenance, preventing hardware failures,” he explained. In cybersecurity, its capabilities can detect anomalies, isolate affected systems, and trigger immutable backups almost instantly, thereby minimizing potential damage.

Skills and Leadership: Preparing for Change

The adoption of Agentic AI will necessitate new human skills. “The role of humans will shift from execution to oversight and orchestration,” Hardy remarked. Leaders must establish boundaries and monitor autonomous systems to ensure they operate within ethical and organizational limits.

Future Implications: Balancing Autonomy with Oversight

The findings from Capgemini and insights from Hardy converge on a central theme: Agentic AI presents immense promise, but its practical impact hinges on balancing autonomy with human oversight. While the technology can help reduce costs and improve reliability, a focus on governance, reskilling, and infrastructure readiness is essential for successful adoption.

Conclusion: The Path Ahead for Agentic AI

As we look toward the future, the critical question remains: How quickly will Agentic AI take hold, and can enterprises effectively balance autonomy with accountability as machines assume greater responsibility for business decisions? The journey is just beginning, and the potential for transformation is significant.

Engaging Questions and Answers

1. What is Agentic AI?

Agentic AI refers to software that can autonomously make decisions, take actions, and refine its strategies based on real-time data and learning. It operates in dynamic environments to pursue specific objectives.

2. What are the economic implications of Agentic AI?

Capgemini estimates that Agentic AI could unlock up to US$450 billion in economic value by 2028, highlighting its potential to significantly impact various industries.

3. Why is trust in AI agents declining?

Trust in AI agents is diminishing due to limited deployment, concerns over data security, and the need for human oversight in AI workflows.

4. How can organizations ensure successful adoption of Agentic AI?

Organizations can ensure successful adoption by focusing on data quality, building supporting infrastructure, and prioritizing governance and reskilling initiatives.

5. What skills will be essential in a world with Agentic AI?

Essential skills will include oversight and orchestration of AI systems, governance capabilities, and a focus on mentoring and strategic thinking rather than administrative tasks.