Setting sail for the future: Agentic AI charts a new course for business

I’m a sailor at heart, and one of my favorite pastimes is taking to the seas with a group of friends. On a recent sailing trip, I had the opportunity to introduce my companions to a new crew member — a ChatGPT-based AI model that I’d been teaching to speak with an Australian accent. It proved to be an invaluable addition to our voyage.

We were a group of five, and none of the others were in the technology industry. I wanted to show them the incredible potential of agentic AI, which is designed to operate with a high degree of independence, handle complex tasks, learn from its environment and adapt to new situations in real time.

To my amazement, our AI crew member managed everything from keeping the required naval log to navigating our sailing trip with precision. It even suggested recipes based on the simple food supplies we had in the fridge — and the meals were delicious!

The experience demonstrated in a very practical way how AI (and, by extension, agentic AI) can make many activities and processes safer and more productive and enjoyable, even in the most unexpected settings.

Meaningful leadership is essential to agentic AI

I shared this anecdote on stage recently during a panel discussion at the World Economic Forum’s Annual Meeting of the New Champions 2025 in Tianjin, China, where we delved into the transformative potential of agentic AI across industries.

One of my fellow panelists, Kian Katanforoosh, CEO and Founder of the skills intelligence platform Workera, emphasized the importance of rapid adoption and adaptation, as organizations that can quickly implement agentic AI will gain a competitive edge.

And Vu Van, Cofounder and CEO of Elsa, which offers AI-enabled speech assistance, highlighted the role of hyperpersonalization, which is crucial in AI-driven customer engagement and in keeping agentic AI solutions both accurate and relevant.

Watch the full panel discussion at the Annual Meeting of the New Champions 2025.

From my perspective, as CEO of NTT DATA Asia Pacific, I’ve seen firsthand how agentic AI is rapidly changing the way we do business on many levels, with great potential for ethical governance and sustainability if we commit to the goal of responsible innovation.

Agentic AI is catalyzing a shift from capital-intensive digital transformation projects to operating models built around assured outcomes. It enables systems to take initiative, make goal-based decisions in real time, collaborate with humans in new ways and help us make more informed decisions.

At NTT DATA, we’re positioning ourselves as a global leader in this space with a full-service, responsible innovation approach, and our newly introduced Smart AI Agent™ Ecosystem supports this transformation.

However, with great potential comes great responsibility. Guanchun Wang, Chairman and CEO of Laiye, which focuses on smart automation platforms, spoke about a future where large organizations will have more digital workers than human ones. The potential for increased productivity and efficiency is undeniable — but the shift will require meaningful leadership and cultural changes.

• ALSO READ → The new frontier: Why agentic AI should be on every CIO’s roadmap

How NTT DATA is deploying agentic AI and improving supply chains

At NTT DATA, we have solutions designed to address the two main approaches to deploying agentic AI in enterprises.

First, many organizations are already using robotic process automation (RPA) or some other form of automation for repetitive, consistent processes. We’ve built a smart AI platform that allows these organizations to integrate an AI agent into their existing RPA environments.

We’ve already done this within NTT DATA, applying agentic AI to back-office processes in finance, HR and other areas where we’ve been using RPA for years. The ability to swap in an autonomous agent has been truly beneficial.

The other scenario involves redesigning complex processes that require a high level of human intervention. A shift manager at a factory might typically plan weekly shifts based on work orders. But if many staff don’t show up after a big football game the previous night, an AI agent can now automatically hire temporary workers and coordinate with suppliers. This requires completely changing the existing processes.

• ALSO READ → The autonomous enterprise: Are you ready for what’s next?

We’re also building specialized agentic AI solutions for supply chain optimization by embedding domain-specific autonomous supply chain technology in our Smart AI AgentÔ Ecosystem. These agents make proactive decisions based on real-time signals, reduce lead times for short turnaround processes, resolve exceptions without human intervention and adapt dynamically to changing market conditions or logistics.

This demonstrates how these AI agents are moving beyond simple automation to enable resilient, data-driven operations and transform fragmented supply chains into intelligent, self-optimizing networks.

The importance of policies and frameworks

A word of caution: while it’s important to deploy agentic AI quickly, rushing without a solid AI usage policy and clear governance framework can lead to significant consequences.

At NTT DATA, all our employees undergo foundational AI training, covering regulatory, ethical and governance aspects, before they start any projects, and we help our clients do the same for their workforces.

We prioritize setting up strong governance and usage policies from the start. These should, for example, focus on auditability and traceability, because it’s crucial to understand and verify the decision-making process, not just performance.

Also, it’s paramount to ensure AI doesn’t harm organizations or society more broadly. This involves educating the workforce and involving them in discussions about how agentic AI can assist them, rather than replace them.

The potential of this technology is immense, but the consequences of mishandling it can be severe for organizations, brands and society. My advice: move ahead confidently but carefully.

Prepare for obstacles along the way

NTT DATA surveyed over 2,300 senior decision-makers globally to understand their investment plans for GenAI. It’s no surprise that 99% expect to invest more in this area over the next two years.

While the potential of AI — and agentic AI in particular — is vast, the challenges are equally significant. During our panel discussion, we spoke about several key obstacles to address:

1. Infrastructure improvements: Agentic AI requires robust infrastructure to support its operations. This includes not only the hardware and software but also the data pipelines and storage solutions that organizations need for the technology to function effectively.
2. AI usage and governance policies: As I already mentioned, there is a critical need for clear and established policies that govern the use of AI. At NTT DATA, we’re developing these frameworks in order to deploy agentic AI solutions responsibly both within our own organization and for our clients.
3. Cybersecurity risks: As agentic AI becomes more integrated into our systems, the risk of cyberthreats increases. We need to invest in advanced cybersecurity measures to protect our digital assets and maintain the trust of our customers and stakeholders.
4. Data readiness: Using agentic AI to deliver hyperpersonalized experiences to customers and employees alike requires sophisticated data management and analytics capabilities. Determining an organization’s data readiness is an essential first step on the road to agentic AI success.
5. Management of digital workers: The rise of digital workers will necessitate changes in organizational structures and management practices. As Guanchun pointed out, leaders will need to develop new skills to manage both human and digital teams — which includes fostering a culture of collaboration and continuous learning in their organizations.

For industries that are heavily regulated, such as finance and healthcare, the adoption of agentic AI comes with additional considerations. These industries must ensure that AI systems are consistently accurate and that their decision-making processes are transparent and verifiable.

A collective responsibility to make data centers sustainable

Agentic AI models consume vast amounts of energy. NTT DATA is the third-largest data center provider globally, and we’re expanding our capacity significantly in the Asia Pacific region. For instance, we operate 14 data centers in India and are making substantial investments to grow even further. Sustainability is therefore a huge priority for us.

We’re exploring and implementing several technologies to reduce our environmental impact, such as liquid cooling and direct-contact cooling. These advances, along with using renewable energy sources like solar and wind, are being integrated into our Asia Pacific data centers.

Our data shows that these technologies can reduce energy consumption by 30%. However, our goal is to achieve net-zero emissions by 2030, and we are fully committed to this target.

It’s the collective responsibility of both organizations and governments to find sustainable solutions for using this technology. We must work together to ensure that the benefits of agentic AI do not come at an unacceptable environmental cost.

Broadening the benefits of AI

The future of the Asia Pacific economy is closely tied to the successful adoption and integration of agentic AI.

However, over 2.5 billion people worldwide still lack internet access. This highlights the need for inclusivity in our AI initiatives. It’s a concern that weighs heavily on me, and it’s why NTT DATA is participating in the World Economic Forum’s Blueprint for Equity and Inclusion in Artificial Intelligence, an initiative to identify gap areas and opportunities to make AI more equitable and inclusive for all. We can’t afford to leave so many people behind in this technological revolution.

Leaders across industries must therefore embrace this technology with a clear vision and a commitment to inclusivity and responsible innovation. By doing so, we can create a more productive, efficient and sustainable future for all.

• ALSO READ → Smart agents, smarter business: Experience the power of agentic AI