Scaling GenAI and AI Agents Across the Enterprise: Ai4 Panel Recap

At Ai4 Las Vegas 2025, Sinequa by ChapsVision joined leaders from TheLoops, Dice, Worth Media, and UCLA to talk about one of the biggest challenges in tech today: scaling GenAI and AI agents across the enterprise.

Panelists included:

• Somya Kapoor, CEO & Co-Founder, TheLoops
• Paul Farnsworth, President, Dice
• Dan Costa, Editor-in-Chief, Worth Media
• Howard Miller, CIO, UCLA Anderson School of Management
• Jeff Evernham, Chief Product Officer, Sinequa by ChapsVision

The conversation captured both the excitement and the hard truths of bringing AI to life inside large organizations. Adoption is accelerating, but so are the lessons learned. While adoption is growing rapidly, the rise in abandoned projects likely reflects the natural progression of the market with many early experiments from last year now reaching the point of evaluation and course correction.

According to S&P Global, 42% of enterprises had abandoned most AI projects by the end of 2024, up from just 17% the year before. So, what separates the success stories from the stalled ones? Here’s what the panel had to say.

Change Management Matters as Much as Technology

Scaling AI isn’t just a technical challenge: it’s an organizational one. The companies seeing the best results have leadership driving adoption from the top and champions pushing progress within departments.

Those that pair AI initiatives with structured change management programs report tangible gains: time saved, smoother operations, and tighter alignment between AI projects and business goals.

Start Small: Focus on Quick Wins, Not “Boil the Ocean” Projects

It’s tempting to aim high with AI, but many early efforts have collapsed under their own weight. The panel agreed: start with focused, high-impact projects that deliver visible results quickly.

For example:

• Customer support teams are using AI to offload repetitive tasks and give agents more time to help customers.
• Universities are piloting AI agents that simulate classroom scenarios, giving every student a chance to participate.
• Product and marketing teams are leveraging AI-powered research and knowledge retrieval tools to move faster from ideas to insights.

Quick, meaningful wins build confidence and momentum and make it easier to scale up successfully later.

Why So Many AI Projects Struggle at Scale

Most organizations can run a successful proof of concept (POC). The challenge comes when moving from limited data sets to enterprise-wide deployments.

At enterprise scale, challenges like security, governance, and data quality take center stage. Without strong foundations, even the smartest model will struggle, it’s the classic “garbage in, garbage out” problem.

RAG Leads the Way

A recurring theme was the role of retrieval-augmented generation (RAG). These approaches ground AI in enterprise content, reduce hallucinations, and provide traceable citations. This is essential for deploying reliable generative AI in the enterprise, because it’s the only way for models to use the knowledge of your business. The common theme: context is crucial for reliable, enterprise-grade AI.

AI Agents: From Helpful Assistants to Autonomous Systems (Eventually)

While the industry buzzes about “autonomous agents,” most companies are still in the early innings. Today’s agents work best as assistants, helping employees get more done, not replacing them.

Panelists warned that trying to deploy too many agents at once leads to confusion and failure. The smarter path? Pilot one use case at a time, prove its value, and expand gradually. Over time, those agents can start interacting with each other, but that level of complexity comes later.

Adoption Depends on People as Much as Technology

Even the best-designed system will fail if people don’t use it. Many employees are understandably wary of AI, worried about job loss or loss of control.

Panelists shared several ways to build trust and drive adoption:

• Frame AI as a helper, not a replacement.
• Let employees customize their AI assistants so the tools fit their workflow.
• Highlight wins early and often success stories spread.
• Recognize internal champions who lead by example.

When employees feel ownership and see real results, adoption follows naturally.

Looking Ahead: Governance, Guardrails, and Strategic Growth

The panel’s closing predictions painted a realistic but optimistic picture for 2025 and beyond:

• Expect more experimentation and more course corrections, especially for large-scale agent projects without clear governance.
• Well-defined, narrow use cases will keep delivering strong ROI.
• Governance, error tracking, and transparency will shift from “nice to have” to “must have.”
• And “scaling” will take on new meaning, not just deploying everywhere, but achieving real business impact with the right focus and teams.

Sinequa by ChapsVision: Powering Scalable, Trusted AI Across the Enterprise

Sinequa by ChapsVision helps enterprises transform the insights from the Ai4 panel into real-world results. Our platform combines RAG and domain-specific AI models to deliver context-aware AI agents that employees trust and adopt quickly.

With Sinequa by ChapsVision, organizations can:

• Accelerate Adoption with Quick Wins: Identify high-impact, focused use cases that demonstrate immediate value and build internal momentum.
• Ensure Trust and Reliability: Deploy AI agents supported by secure, governed, and traceable access to enterprise data, minimizing hallucinations and maximizing confidence.
• Scale Strategically: Start with targeted pilots, then expand across teams and business functions, all while maintaining control over data, compliance, and performance metrics.
• Empower Employees: Enable employees to interact with AI agents tailored to their roles, turning the technology into a productivity co-pilot rather than a replacement.

By combining AI expertise with enterprise-grade infrastructure and change management support, Sinequa by ChapsVision helps companies turn the promise of GenAI and AI agents into measurable business impact, faster, safer, and smarter.

Discover how Sinequa by ChapsVision can help your organization scale AI successfully and unlock the full potential of your knowledge.