The use of this site and the content contained therein is governed by the Terms of Use. When you use this site you acknowledge that you have read the Terms of Use and that you accept and will be bound by the terms hereof and such terms as may be modified from time to time.
All text, graphics, audio, design and other works on the site are the copyrighted works of nasscom unless otherwise indicated. All rights reserved.
Content on the site is for personal use only and may be downloaded provided the material is kept intact and there is no violation of the copyrights, trademarks, and other proprietary rights. Any alteration of the material or use of the material contained in the site for any other purpose is a violation of the copyright of nasscom and / or its affiliates or associates or of its third-party information providers. This material cannot be copied, reproduced, republished, uploaded, posted, transmitted or distributed in any way for non-personal use without obtaining the prior permission from nasscom.
The nasscom Members login is for the reference of only registered nasscom Member Companies.
nasscom reserves the right to modify the terms of use of any service without any liability. nasscom reserves the right to take all measures necessary to prevent access to any service or termination of service if the terms of use are not complied with or are contravened or there is any violation of copyright, trademark or other proprietary right.
From time to time nasscom may supplement these terms of use with additional terms pertaining to specific content (additional terms). Such additional terms are hereby incorporated by reference into these Terms of Use.
Disclaimer
The Company information provided on the nasscom web site is as per data collected by companies. nasscom is not liable on the authenticity of such data.
nasscom has exercised due diligence in checking the correctness and authenticity of the information contained in the site, but nasscom or any of its affiliates or associates or employees shall not be in any way responsible for any loss or damage that may arise to any person from any inadvertent error in the information contained in this site. The information from or through this site is provided "as is" and all warranties express or implied of any kind, regarding any matter pertaining to any service or channel, including without limitation the implied warranties of merchantability, fitness for a particular purpose, and non-infringement are disclaimed. nasscom and its affiliates and associates shall not be liable, at any time, for any failure of performance, error, omission, interruption, deletion, defect, delay in operation or transmission, computer virus, communications line failure, theft or destruction or unauthorised access to, alteration of, or use of information contained on the site. No representations, warranties or guarantees whatsoever are made as to the accuracy, adequacy, reliability, completeness, suitability or applicability of the information to a particular situation.
nasscom or its affiliates or associates or its employees do not provide any judgments or warranty in respect of the authenticity or correctness of the content of other services or sites to which links are provided. A link to another service or site is not an endorsement of any products or services on such site or the site.
The content provided is for information purposes alone and does not substitute for specific advice whether investment, legal, taxation or otherwise. nasscom disclaims all liability for damages caused by use of content on the site.
All responsibility and liability for any damages caused by downloading of any data is disclaimed.
nasscom reserves the right to modify, suspend / cancel, or discontinue any or all sections, or service at any time without notice.
For any grievances under the Information Technology Act 2000, please get in touch with Grievance Officer, Mr. Anirban Mandal at data-query@nasscom.in.
AI agents, also known as intelligent agents, are software programs designed to perceive their environment, take actions, and achieve specific goals autonomously. They differ from traditional computer programs in their ability to learn and adapt, make decisions, interact with surroundings, and operate with limited supervision. Agentic AI systems integrate one or more AI agents that collaborate with each other and provide a unified seamless experience and outcome for the end user.
Chatbots have been in existence for a while. Are chatbots AI agents? Or do they differ from AI agents?
AI agents and chatbots share similarities in their ability to interact with users, but differ significantly in their capabilities and underlying technologies:
Chatbots
AI Agents
Focus
Predefined tasks, scripted responses, and simple interactions
Complex tasks, dynamic decision-making, and collaboration with other agents
Technology
Often rule-based, relying on keyword matching and pre-programmed responses
Built on AI techniques like machine learning, natural language processing, natural language understanding, knowledge representation, and basic levels of causal reasoning
Capabilities
Follow scripts: Provide predefined responses based on user input, keywords, or decision rules
Limited adaptivity: Cannot learn or adapt significantly beyond their initial programming
Independent: Function individually and typically do not collaborate with other agents.
Learn and adapt: Can continuously learn from data, user interactions, and improve their performance over time
Reason and plan: Can understand context, reason through problems, and make informed decisions based on their goals
Collaborate: Can work together with other agents to achieve complex goals that require coordination and communication
Example
An example of a chatbot is that of a customer service bot that uses a predefined script, provides basic information, and directs you to appropriate resources based on predefined options.
An example of an AI Agent is a highly trained personal assistant who can understand your needs, learn your preferences, and take initiative in helping you achieve your goals.
In essence, AI agents are like general-purpose tools that can be adapted to various tasks requiring intelligence, decision-making, and collaboration. Chatbots, in contrast, are like specialized tools that excel at specific, well-defined tasks but lack the flexibility and adaptability of AI agents.
How to build Agentic AI systems?
AI Agent Frameworks for Building Collaborative Intelligence
AI agent frameworks are a combination of libraries and workflows that facilitate the creation and management of intelligent software agents. Following are some of the latest frameworks to build AI agents:
AutoGen
AutoGen from Microsoft provides a multi-agent conversation framework as a high-level abstraction. It is an open-source library for enabling next-generation LLM applications where users can build LLM workflows with multi-agent collaborations and personalization. The agent modularity and conversation-based programming simplify development and enable reuse for developers.
Use case: An enterprise knowledge management system with a conversational interface using a knowledge base agent, retrieval agent, and dialogue agent.
CrewAI
CrewAI is an open-source framework built on top of LangChain for creating and managing collaborative AI agents. It enables developers to build cohorts of specialized AI agents that can work together to achieve complex tasks.
Use Case: A marketing team could use CrewAI to create a series of agents – one to gather customer data from social media, another to analyze sentiment, and a third to generate targeted marketing campaigns based on the insights.
LangGraph
LangGraph is also another open-source framework built on top of LangChain. It helps represent multiple agents in a graph network and ensures seamless integration and collaboration.
Use Case: An AI Research assistant that comprises a research content web scraping agent, a processing agent that identifies relevant content default behavior and synthesizes and stores curated content, and a generating agent that crafts initial drafts of research papers based on user goals and objectives
Challenges of Adopting Agentic AI
Despite the significant advancements in Agentic AI, there are several key challenges that still need to be addressed, such as:
Unforeseen consequences: Agentic AI systems, due to their adaptability and ability to learn, can potentially engage in unforeseen actions or decisions, leading to unintended consequences.
Limited understanding of internal workings: The complex decision-making processes within these systems can be opaque. This can make it difficult to identify the root cause of errors or failures.
Transparency in data usage and processing: Concerns exist regarding potential misuse of user data by agentic AI systems and the need for transparent practices in data collection, storage, and utilization.
Unmitigated bias: Training data and algorithms can contain inherent biases that agentic AI systems may learn and perpetuate, leading to discriminatory or harmful outcomes.
Understanding decision-making: It’s often challenging to understand how agentic AI systems arrive at specific decisions, hindering user trust and hampering troubleshooting or improvement efforts.
Guide to Successful Development & Implementation of Agentic AI
As agentic AI systems get mainstream with their ability to accomplish complex goals, there is a need for a robust governance framework to overcome the challenges. A recent paper from OpenAI titled “Practices for Governing Agentic AI Systems” outlines some guidelines for safe and responsible development and deployment of such systems. Following are some key insights from the paper that would enable the responsible development and adoption of such systems:
Defining Responsibilities:
Clear roles and liabilities: Clearly define who is responsible for the actions of agentic AI systems throughout their lifecycle, including developers, deployers, and users. This promotes accountability and mitigates potential harm.
Attributability: Provide a unique identifier to AI agents so that it is possible to trace the source of error when required.
Ensuring Safety:
Robust safety measures: Implement safeguards like regular audits, human oversight for critical decisions, and clear guidelines for acceptable actions to minimize potential risks and unintended consequences.
Constrain the action space and seek approval: In some cases, prevent agents from taking specific actions entirely to ensure safe operation. It is prudent to have human-in-the-loop for review and approval when the cost of wrong decisions and actions can be catastrophic.
Timeouts: Implement mechanisms to periodically pause the agent operation and require human review and reauthorization, preventing unintended harm from continuous unsupervised operation.
Setting the Agent’s default behavior: Reduce the likelihood of the agentic system causing accidental harm by proactively shaping the model’s default behavior that reiterates user preferences and goals to steer toward actions that are the least disruptive ones possible, while still achieving the agent’s goal.
Transparency and Explainability: Ensure the reasoning and decision-making processes of agentic AI systems are clear and understandable to the extent possible. This fosters trust and allows for identification of potential biases or flaws.
Automatic Monitoring: Set up a Monitoring AI system that automatically reviews the primary agentic system’s reasoning and actions to check that they are in line with the user’s goals and expectations.
Ad hoc Interruption and Maintaining User Control: User should always be able to activate a graceful shutdown procedure for its agent at any time, both for halting a specific category of actions and for terminating the agent’s operation more generally.
Ethical Considerations: Ensure that the development and deployment of agentic AI systems adhere to ethical principles and societal values. This includes promoting fairness, non-discrimination, privacy, and overall human well-being.
Public Dialogue and Participation: Encourage open discussions and collaboration between experts, policymakers, and the public to shape responsible AI development and ensure it aligns with societal values.
It’s important to note that OpenAI’s framework is just a starting point, and ongoing research and discussion are crucial in developing comprehensive and effective governance models for agentic AI systems.
The Road Ahead
Real-world systems involve amalgamation of multiple capabilities, which warrants the design of Agentic AI systems that use multiple AI agents. We are seeing the emergence of such design patterns, given the limitations of large language models (LLMs) in producing outputs with just one API call for complex tasks. Since the quality of output of a system is multiplicative of the individual output quality from each subsystem, each subsystem would need its own output verification, validation, and feedback loop to ensure reliable and trustworthy outcomes. Governance of agentic AI is an ongoing process that requires continuous adaptation and improvement as the technology evolves.
By fostering collaboration, promoting transparency, and prioritizing ethical considerations, we can navigate the development and deployment of agentic AI responsibly and reap its benefits for the betterment of society. Agentic AI systems offer immense potential and are going to be a game changer in the coming days.
That the contents of third-party articles/blogs published here on the website, and the interpretation of all information in the article/blogs such as data, maps, numbers, opinions etc. displayed in the article/blogs and views or the opinions expressed within the content are solely of the author's; and do not reflect the opinions and beliefs of NASSCOM or its affiliates in any manner. NASSCOM does not take any liability w.r.t. content in any manner and will not be liable in any manner whatsoever for any kind of liability arising out of any act, error or omission. The contents of third-party article/blogs published, are provided solely as convenience; and the presence of these articles/blogs should not, under any circumstances, be considered as an endorsement of the contents by NASSCOM in any manner; and if you chose to access these articles/blogs , you do so at your own risk.
C5i is a pure-play AI & Analytics provider that combines the power of human perspective with AI technology to deliver trustworthy intelligence. The company drives value through a comprehensive solution set, integrating multifunctional teams that have technical and business domain expertise with a robust suite of products, solutions, and accelerators tailored for various horizontal and industry-specific use cases. At the core, C5i’s focus is to deliver business impact at speed and scale by driving adoption of AI-assisted decision-making. C5i caters to some of the world’s largest enterprises, including many Fortune 500 companies. The company’s clients span Technology, Media, and Telecom (TMT), Pharma & Lifesciences, CPG, Retail, Banking, and other sectors. C5i has been recognized by leading industry analysts like Gartner and Forrester for its Analytics and AI capabilities and proprietary AI-based platforms.
MLOps is quickly becoming one of the most popular technologies in the field of machine learning. This is due to its ability to provide an efficient and effective way to manage, automate, and deploy machine learning models in production environments…
Preventative medicine is all set to make a comeback as hospitals now have the tools that are required to collect, analyze and deliver solutions that map the trajectories of their patient’s health in a sustainable fashion. Telemedicine, as the…
Organizational experts know that the best outcomes are achieved by building a collaborative team environment, one where information is shared and team members work from a common platform of goals and data so that everyone is on the same page.…
In the past 10 years, the term ‘disruption’ has been abuzz across business and industry circles. Whether succeeding as a disruptive innovator or defending against a challenger as an incumbent, the nature of innovation has focused on the ability to…
As businesses plan for data democratization, it is important to include a strategy to ensure that business users will accept augmented analytics solutions and adapt to the new Citizen Data Scientist role. With business users on board, the enterprise…
When a business considers the implementation of an analytics solution, it may be focused on choosing the easiest, most flexible analytics solution to encourage its users to make analytics part of the process and to incorporate it into everyday tasks…
