Artificial Intelligence & Machine Learning

Agentic AI: The Next Frontier for Enterprises

Generative AI has captured the attention of businesses and tech leaders in recent years. Among its most significant advancements is agentic AI. Unlike traditional chatbots or rule-based systems, agentic AI doesn't just respond; it learns, reasons and acts autonomously, promising to elevate how companies interact with customers and manage workflows.

Gartner predicts that by 2028, at least 15% of day-to-day work decisions will be made autonomously through agentic AI. Industries that demand high levels of contextual accuracy, operational efficiency and seamless customer experiences are seeing agentic AI emerge as a transformative force. By integrating vast datasets, using machine learning models and acting in real time, these intelligent systems are setting new benchmarks for innovation.

From Scripted Bots to Intelligent Agents

For years, enterprises have relied on traditional chatbots to handle routine queries. While effective for simple tasks, these chatbots often struggle with personalization, nuanced responses and real-time adaptability. Their reliance on heavily scripted interactions limits their flexibility and utility, especially in dynamic environments.

Agentic AI represents a significant leap forward. "These systems can perform complex reasoning, integrate with vast enterprise datasets and execute processes autonomously. For instance, a task like merging customer accounts, which traditionally required ticket creation and days of manual effort, can now be completed in seconds with agentic AI," said Arun Kumar Parameswaran, senior vice president and managing director of Salesforce India.

The capability to deliver autonomous, end-to-end solutions makes agentic AI indispensable in addressing the growing demands of modern businesses. Valued at $5.1 billion in 2024, the market for agentic AI is experiencing rapid growth, projected to reach $47.1 billion by 2030, growing at a CAGR of 44.8%. Salesforce and Oracle are among the key players driving this momentum, with both companies announcing new AI agent platforms that cater to diverse enterprise needs.

Agentic AI can autonomously generate and manage documents, set up profiles, and configure IT access, streamlining traditionally manual processes.

Agentforce: Leading the Charge

Salesforce's Agentforce, unveiled at Dreamforce 2024, represents a significant milestone. Built on the company's Atlas reasoning engine and using models such as OpenAI's GPT-4 and Google's Gemini, Agentforce combines advanced AI with Salesforce's extensive ecosystem of customer engagement data.

Agentforce marks the "third wave of AI," said Marc Benioff, CEO of Salesforce. He predicts a massive 1 billion AI agents by 2026. Unlike earlier waves, which focused on predictive analytics and conversational bots, this phase emphasizes intelligent agents capable of autonomous decision-making. Salesforce has amassed years of customer engagement data, workflows and metadata, making Agentforce a precision tool that understands and anticipates customer needs.

Agentforce is built to democratize AI adoption. Its no-code approach allows businesses, including small and medium enterprises, to deploy autonomous agents without requiring technical expertise. This accessibility ensures that organizations of all sizes can benefit from AI-driven innovation. Salesforce is also exploring a consumption-based pricing model, with costs estimated at $2 per conversation, making the technology scalable and cost-effective.

Real-World Applications

Early adopters demonstrate agentic AI's potential to transform operations.

• Wiley, a global leader in research and learning, is driving tangible improvements by integrating Agentforce with its existing systems. It has reduced onboarding time for seasonal staff by 50%, enabling the company to scale its workforce more effectively during peak periods. Wiley has reported a 213% return on investment and $230,000 in savings. The platform's ability to draft accurate responses using Wiley's CRM data and knowledge base has significantly enhanced agent productivity. Automating routine tasks, such as resetting passwords and resolving account issues, has also allowed human agents to focus more on complex customer needs. This strategic shift has improved case resolution rates by over 40% within just a few weeks of implementation.
• Premier luxury retailer Saks uses Agentforce to handle routine inquiries, such as order statuses and returns, freeing up its team to deliver personalized, high-touch customer experiences. By using first-party data and AI, Saks is creating a unified shopping experience across its digital, mobile and in-store channels.
• OpenTable, a leader in restaurant technology, uses Agentforce to automate tasks such as reservation changes and loyalty point redemptions, enabling its support team to focus more on complex customer interactions. Data integration is a cornerstone of OpenTable's success with Agentforce. By integrating data from multiple sources, the company has created a single platform that provides a comprehensive view of each diner, including their preferences and past interactions. This holistic approach has allowed OpenTable to deliver highly personalized experiences across multiple touchpoints, from email to SMS and push notifications. The company utilized natural language prompts and intuitive configurations to launch the system seamlessly.

Challenges and Considerations

Despite its immense potential, agentic AI is not without challenges. A recent Gartner report estimates that at least 30% of gen AI projects will fail to progress beyond the proof-of-concept stage by 2025. Key hurdles include poor data quality, unclear business objectives and high implementation costs.

For enterprises looking to adopt agentic AI, having clean, organized data is critical. Without this foundation data, the results of AI deployments may fall short of expectations. Additionally, building AI systems in-house can be prohibitively expensive and resource-intensive.

"Organizations must ensure that these systems are designed and implemented securely, with robust governance frameworks to mitigate potential risks," said Deepak Pargaonkar, vice president of solution engineering at Salesforce.