Trusted Knowledge Is Key to AI Success, eGain CEO Asserts at Solve 25

CHICAGO -- Using trusted knowledge is the key to companies driving success with artificial intelligence, Ashu Roy, eGain chairman and CEO, told the audience during his opening keynote at the company's Solve 25 conference.

There is no doubt about the popularity of AI, Roy said, pointing to the 800 million users of OpenAI's ChatGPT and the 1 million downloads of OpenAI's Sora app in less than five days, as well as the expected impact of the technology in replacing human jobs.

In addition to providing better efficiency, the technology continues to evolve and become less expensive, with capabilities doubling about every six months.

Yet, despite the popularity and the several millions of dollars of investment in the technology, only 5 percent of users are showing a positive ROI, according to Roy.

The problem, he said, is that too many businesses are trying to apply the technology in inefficient, siloed technology stacks. "Companies need to know how to build things to take advantage of AI's capabilities."

The way to build AI-based technology, which can be particularly helpful in contact centers and other customer experience functions, is to look at the solutions as helping provide many small solutions, each of which drives incremental value.

There will still be exceptions the technology can't answer, Roy said. Those exceptions will be handled by a human in the loop.

But even with a human in the loop, users have to trust AI is providing trusted knowledge and not hallucinating, Roy says. eGain provides this trusted knowledge with technology that includes the following:

• Intelligent discovery: The AI automatically analyzes conversations and questions from customers and employees to identify the knowledge that truly matters.
• Content cleansing: The AI helps eliminate the drudgery of identifying and eliminating duplicate or conflicting content.
• Automated curation: Trusted enterprise knowledge content is sourced, connected, and curated automatically by AI in seconds, with AI prompts ensuring adherence to templates and style guides.
• Deterministic reasoning: The AI automatically converts compliance-heavy procedures to deterministic reasoning.
• Continuous Optimization: Ongoing maintenance and publishing across the enterprise is automated, including content taxonomies, organization, and business-aligned metadata generation.

The keynote was followed by a more in-depth discussion of eGain AI Agent 2, which the company introduced at the conference along with eGain Composer.

AI Agent 2 provides users with a combination of trusted knowledge and hybrid AI, enabling users to overcome the obstacles of incomplete knowledge bases, inconsistent answers, and unreliable handling of complex, multi-step processes.

The product provides trusted knowledge and ensures that agentic interactions are grounded in accurate, up-to-date information. It delivers Assured Actions through a combination of hybrid AI reasoning and quality assurance capabilities built on the eGain AI Knowledge Hub.

Among AI Agent 2's features are the following:

• Probabilistic reasoning from large language model.
• Deterministic reasoning for specific, multi-step workflows.
• The PrismEval Service, which continuously optimizes the match between the knowledge base and answers delivered by AI.
• Assured actions for consistent experiences in the same context every time.
• Omnichannel support across phone, email, chat, messaging, and social media.
• Dual deployment options for contact center agents and customer self-service.
• Pre-built integrations with leading platforms, including Amazon Connect, Genesys, Salesforce, and Talkdesk.

eGain Composer, a new modular AI knowledge platform, helps developers integrate the eGain AI Knowledge Hub into their CX applications.

Today's AI knowledge systems--standalone knowledge bases, retrieval engines, agentic interfaces, and experience design frameworks--are often fragmented, disconnected, and inflexible, according to Roy, who added that Composer addresses these challenges by enabling developers to build trusted AI CX applications at scale.

Among the benefits of Composer, beyond the integration with the Knowledge Hub platform, are the following:

• Content lifecycle management: Developers have granular API controls throughout the content management lifecycle, from authoring to deprecation.
• Advanced compliance with security and authentication standards, including OAuth 2.0, HTTPS, SOC 2, HIPAA, GDPR, and FedRAMP.
• Agentic readiness for assured actions: Seamless integration with AI agents provides enterprise search, answers, and agentic actions across customer-facing platforms.
• Composable architecture: The APIs enable flexible, plug-and-play integrations for web, desktop, mobile, and agentic environments.
• Content health: Closed-loop optimization cycles ensure ongoing improvement in search effectiveness and agentic reliability, enabling developers to implement quality-driven workflows.

eGain Composer delivers these benefits through robust application programming interfaces (APIs), Model Context Protocol (MCP) servers, and software development kits (SDKs) in Python and TypeScript. Through APIs and the MCP server, eGain Composer surfaces trusted enterprise knowledge in diverse formats, from search results and AI-curated answers to agentic actions and embedded integrations. The platform is a comprehensive tool for synthesizing and curating content, managing prompts, and integrating third-party services. MCP servers, available on GitHub in both local and remote configurations, support ingestion, search, answer delivery, and agentic Assured Actions. Using OpenAPI specifications, eGain Composer APIs enable flexible integrations across diverse environments.

"Today's enterprises want a full-stack AI knowledge solution while being able to mix and match the building blocks at the same time," Roy said. "eGain Composer allows them to do just that.">

eGain also noted that companies are looking for ways to capture worker knowledge as Baby Boomers retire. More than 11,000 Americans turn 65 every day, with more than half of the entire workforce planning to retire in the next five years, the company said, citing a survey by AQPC.

"The threat of the Great Retirement is more than a demographic shift. It's a wakeup call for organizations and C-suite leaders to act now," said Lynda Braksiek, AQPC principal research lead for knowledge management, said in a statement. "As our most valued experts retire in record numbers, the risk of losing that expertise grows. By embracing time-tested knowledge transfer methods and enabling them with AI-driven capabilities, organizations can not only preserve their institutional memory but also unlock new levels of productivity and resilience."

Among the AQPC survey's other findings were the following:

• 58 percent of C-suite respondents are very worried about the knowledge loss.
• 92 percent of organizations do not consistently capture knowledge from soon-to-be retirees.
• 85 percent have not operationalized AI to automate knowledge management.

A major part of the solution, according to Roy, is to capture knowledge from real-time conversations and collaboration using AI to automate the know-how elicitation effort at scale.