Select a country Global - English
Contact Us

From cockpit to command center: How agentic AI is redefining business operations

The adoption of agentic AI will separate the leaders from the laggards in the new era of business.

Imagine an AI system as the pilot of a modern aircraft. It navigates altitude, speed and direction, handling complex decisions and routine tasks with precision. Human pilots are still in the cockpit, but their role has shifted: they supervise, guide and intervene only when necessary.

This transformation didn't happen overnight. In the 1920s, pilots flew manually and managed every detail themselves. Over time, automation took over repetitive functions, freeing pilots to focus on the big picture.

AI is driving a similar evolution in business - from manual processes to intelligent assistance and increasingly autonomous operations. Here's how.

Phase 1: Early autopilots = Robotic process automation

If we continue to track the trajectory of AI in terms of the evolution of the aviation industry, we can think of the early pilots who had to do everything manually as akin to workers who used to handle every repetitive digital task themselves - for example, copying data or updating records.

Robotic process automation (RPA) changed all that. RPA software, or "bots", could now follow simple rules to automate repetitive tasks, freeing humans from digital grunt work. Just like early autopilots, RPA runs predefined scripts, but if something unexpected happens, it gets stuck. It can't improvise or adapt to change.

Phase 2: Assisted flight = GenAI copilots

As autopilots grew smarter, they began handling more of the flight tasks, allowing pilots to monitor and guide. This is similar to the current era of AI. Tools like Microsoft 365 Copilot and GitHub Copilot can assist in real time with writing, coding, analyzing and providing help based on context or prompts.

These tools can't replace humans and do not aim to, but they significantly increase productivity and creativity - much like autopilot technologies help pilots fly smoother and safer.

Phase 3: Autonomous flight = Agentic AI

The agentic AI era is like having a plane that can take off, fly, navigate turbulence and land smoothly, all on its own. In business, that means AI isn't just assisting. It's also managing entire workflows, making real decisions, taking action and adjusting in real time.

This stage involves more than just automation - it's about transformation, with AI not just following instructions but also collaborating, learning and taking initiative.

Here are 10 characteristics that define this new era of agentic AI:

  1. AI-native business processes are created from scratch rather than retrofitted into legacy workflows. These processes are built with autonomy in mind.
  2. Multiagent systems collaborate across functions (such as finance, supply chain or marketing) and link insights from across the enterprise to achieve shared goals. For example, it might involve delaying a marketing campaign based on supply chain constraints or adjusting pricing based on customer sentiment and competitor behavior.
  3. Dynamic, self-optimizing workflows allow processes to evolve in real time - responding to data, demand or disruption instantly.
  4. Adaptive learning agents improve as they operate, learning from new data, feedback and patterns without being reprogrammed.
  5. Agentic AI systems are outcome-driven. They consider the high-level objectives (for example, "optimize inventory across regions" or "increase customer retention") and help users determine the best approach, adapting strategies as conditions change.
  6. These systems shift from rule-based to context-based, and they operate with a deep understanding of past experiences. By integrating historical data, real-time inputs and external factors, they make informed, nuanced decisions aligned with the broader business environment.
  7. The human role shifts: Instead of operating tasks, humans become supervisors, orchestrators and ethical overseers who are setting goals, interpreting results and ensuring responsibility.
  8. Governance, transparency and accountability become essential. Enterprises must know what the AI did, why and who it's acting on behalf of - raising the bar on ethical boundaries, compliance checks, explainability, auditability and identity verification.
  9. Natural and intuitive interfaces make it easier for anyone to work with AI through conversation, visual inputs or proactive suggestions. No coding is required.
  10. For agentic AI to be successful, a comprehensive agent-management framework will be required (like talent management for humans) to build, onboard, continuously train (both technical and soft skills like collaboration) and then retire agents.

Are you ready to fly?

We will see the rise of roles such as Chief Agentic Officers who, alongside their teams, will orchestrate these events to ensure that agentic AI results in overall gains in productivity and brings clarity and growth instead of chaos and confusion.

This enterprise-wide transformation is inevitable, yet too many organizations remain stuck - paralyzed by uncertainty, distracted by complexity or misaligned on the true strategic value of AI. Others have made initial moves but stalled at scale, slowed by fragmented strategies, reactive thinking or a narrow focus on short-term ROI with no unifying vision.

The journey from copilot to autonomy is well underway. The question remains: Where are you on your organization's flight path?

WHAT TO DO NEXT

Read more about NTT DATA's Smart AI Agent™ Ecosystem to see how we can help your organization adopt these technologies effectively and ethically.

Abhijit Dubey

Chief Executive Officer, NTT DATA, Inc.

Abhijit Dubey is the CEO of NTT DATA, Inc. and a member of the advisory board of NTT Venture Capital. Previously, he served as CEO of NTT Ltd., which merged with NTT DATA to form a $30 billion business and technology services powerhouse under the NTT DATA name. Abhijit has over 20 years of experience in the technology and management consulting industries, joining NTT Ltd. in February 2021 from McKinsey & Company, where he was Senior Partner and core leader of the Global Tech, Media and Telecom Practice. Originally from India, he holds an MS in Industrial Engineering from Stanford University and a BTech in Mechanical Engineering from the Indian Institute of Technology, Bombay.