Agentic AI: Technology that will work for the world on its own agency

Agentic AI can process a payment claim on its own or generate images for a social media post without human prompt. It has the agency to learn from its environment

A new era in artificial intelligence (AI) is underway. In a few months, evolved versions of the technology will emerge and find mainstream applications. From AI that works on human prompts, the world will face agentic AI, an autonomous and automated version of the technology. 

 

The impact of such evolving versions of AI will be deep and disruptive. Governments will struggle with the regulatory aspect, while private enterprises will have to be in a constant change mode. Companies will have to be constantly agile and consumers sharp enough to adopt the technology for their personal use.

 

Every dimension of business and governance will face fresh waves of change. Intelligent agents will make AI more useful, according to Gartner. “Today’s AI models perform tasks such as generating text, but these are ‘prompted’ – the AI isn’t acting by itself. That is about to change with agentic AI, or AI with agency.”

 

According to the management consultancy company, about one third of enterprise software applications will include agentic AI by 2028. Agentic AI will enable 15 per cent of day-to-day work decisions to be made autonomously. Today, humans work with AI tools to complete various functions. Direction and objectives are set by humans. Agentic AI would change that. 

 

Chipmaker Nvidia is among the top technology companies betting on agentic AI. The advanced AI system uses sophisticated reasoning and iterative planning to autonomously solve complex, multistep problems, it says. Agentic AI systems ingest vast amounts of data from multiple sources to independently analyse challenges, develop strategies and execute tasks like supply-chain optimisation, cybersecurity vulnerability analysis and helping doctors with time-consuming tasks, according to an assessment by Nvidia. 

 

For example, an agentic AI system may be able to process a payment claim on its own. Another such system will understand the context of a social media post and generate images without human prompt. 

 

The engine for agentic AI are large language models, a type of AI model that has been trained through deep learning algorithms to recognise, generate, translate, and/or summarise vast quantities of written human language and textual data. However, agentic AI does not depend on human instruction and moves autonomously based on certain embedded triggers in the software.

 

"In 2025, the rise of agentic AI, combined with advanced orchestration, is set to transform the technological landscape. This evolution isn’t about machines taking over tasks; it’s about fostering a collaborative ecosystem where AI amplifies human ingenuity,” said Arun Balasubramanian, vice-president and managing director, India and South Asia, UiPath, a software company, in a statement.

 

“We envision a future where agentic AI and agentic automation drive the next wave of innovation. To stay competitive and future-ready, organisations must harness this power, underpinned by strong data governance and security frameworks,” he said.

 

According to Gartner, agentic AI will be incorporated into AI assistants and built into software platforms, internet-of-things devices and robotics. 

 

A fully mature intelligent agent will have the agency to learn from its environment, create complex plans and perform tasks autonomously, said Gartner. 

 

For enterprises, agentic AI will need another wave of organisational and workflow change. UiPath predicts that businesses will begin reimagining roles, workflows, and operational models.  The shift will require significant investments in upskilling, retraining, and reassigning employees.

 

“Orchestration capabilities will become increasingly critical for coordinating tasks, managing workflows, and optimising operations across diverse enterprise technologies and systems,” said UiPath.

 

Business leaders and policymakers will have to track the evolving versions of AI.