Generative AI (GenAI) has proven its ability to generate content based on tailored prompts for specific personas. However, when applied in the context of enterprises, it must be integrated with an organization’s source of truth, a persistent challenge. This source of truth is fragmented across numerous systems, including SharePoint, One Drive, Confluence, Google Drive and many more. As GenAI becomes more widely adopted, the demand for a centralized solution that can seamlessly access these diverse data sources and provide consistent, context-aware responses, all while adhering to enterprise security and governance guardrails, has never been greater.

According to Gartner, by 2028, at least 15% of day-to-day work decisions will be made autonomously through agentic AI. Additionally, one-third of interactions with GenAI services will invoke action models and autonomous agents to complete tasks. This growing shift toward autonomous decision-making and task automation highlights the increasing role of AI in optimizing enterprise operations, further underscoring the need for integrated AI solutions that can navigate complex, fragmented data landscapes.

To address this gap and meet these demands, companies like Google have introduced platforms that enable GenAI-powered search across various data sources, while also providing a foundation for onboarding custom AI agents tailored to specific personas or functions, such as Sales, Marketing, HR and Software engineering.

Google's Agentspace allows for the deployment and management of AI agents, offering significant potential for enterprises across multiple, diverse industries

We envisage that custom agents will generally fall into two categories:

• Custom agents doing unitary task

Consider an HR professional involved in the hiring process who needs to navigate past documents to create job descriptions, formulate interview questions or assess candidates. With cloud-based AI services, custom unitary agents can be created for specific tasks, significantly improving the productivity of hiring managers.

Similarly for a Sales Manager, using custom unitary agents to generate targeted campaigns, perform quick analytics on sales performance and more.

Solutions, like Google’s Agentspace, have simplified access to and the availability of these agents, offering a centralized platform where different users can easily find all the tools and resources they need to improve their effectiveness.

• Custom agents working together to achieve a business outcome

Consider the scenario of a supply chain professional who interacts with multiple different systems to manage inventory, predict demand, negotiate with suppliers and process invoices. To operate effectively, they need to be well trained in the various business applications used to gather information and take relevant actions. Instead of manually navigating these business applications, they could rely on an AI agent that understands the context of their actions and autonomously takes the necessary steps across the backend systems and applications. This approach would not only improve employee productivity but will also reduce the time spent training on separate business applications, leading to reduced operational costs, increased throughput and improved resource utilization.

These custom agents can be deployed in various contexts, depending on the persona and business needs.

While the potential of deploying custom AI agents is immense, enterprises must address several challenges:

• Security and privacy: Ensuring the security and privacy of sensitive data is crucial. Robust authentication, authorization, and encryption mechanisms are essential.
• Ethical considerations: Addressing potential biases and ensuring fairness in agent decision-making is vital. Developing ethical guidelines and frameworks is critical.
• Interoperability and standardization: Establishing standards for agent communication and interaction is necessary to ensure seamless collaboration between different agents and platforms.
• Governance and control: Implementing effective governance mechanisms to manage agent behavior and ensure alignment with business objectives is essential.
• Complexity management: Building and maintaining a complex agent ecosystem requires significant technical expertise and careful planning.

As agentic AI adoption accelerates, the future will revolve around Agentas a Service and its seamless integration into business processes and applications, which should significantly enhance user experience, while removing the complexity of underlying systems.

Solutions like Agentspace represent a paradigm shift in how enterprises will leverage agentic AI. By fostering collaboration, communication and intelligence across multiple AI agents, businesses can unlock unprecedented opportunities for automation, innovation, and personalization. While challenges exist, the potential benefits are transformative, enabling enterprises to achieve new levels of efficiency, agility and competitive advantage. Embracing agentic AI requires a strategic vision, a commitment to ethical development, and a willingness to adapt to the evolving landscape of AI. By proactively addressing the challenges and harnessing the potential, enterprises can unlock the full power of collaborative AI and shape the future of their industries.

At HCLTech we have developed 45+ ready to consume AI agents across various industry verticals, enterprise business functions and software engineering, with over 300 additional agents currently in development based on customer use cases. Additionally, we have also created Agentspace persona specific deployment scenarios and established a well-defined commercial model for agent development and customization to supercharge any agentic journey.

To find out more about how agentic AI can transform your industry, click here: HCLTech Agentic AI solution with Google Cloud | HCLTech