How Enterprises Are Scaling GenAI and Agentic AI: Insights from the Front Line

Posted by: digitalfinanceconference September 18, 2025 No Comments

In a recent presentation, Paul Hewitt, Senior Director, Data & Ai, Strategy and Offerings at DXC—a leading global provider of information technology services shared practical insights into the current state of AI adoption across industries, with a strong emphasis on financial services. The session combined strategic overviews with real-world examples and regulatory considerations, offering valuable guidance for any organization seeking to scale responsibly.

Key Themes Covered:

AI Market Growth: The global AI market is experiencing rapid expansion, with projected annual growth rates between 25–50%. Large Language Models (LLMs) and GenAI dominate investment activity today, especially within financial services, though other AI segments continue to grow in parallel.
Enterprise Maturity Levels: While nearly every organization has experimented with GenAI, most implementations remain limited to chatbot use. Only a small fraction have deployed more complex, production-grade use cases.
Value-Driven Use Cases: Areas with measurable returns include developer productivity (with up to 70% of new code now AI-generated), compliance analytics (e.g., KYC), legacy system modernization, and end-to-end support automation. A standout case involved using Agentic AI to reduce IT support costs by enabling AI agents to collaboratively troubleshoot and resolve complex issues.
Agentic AI: A key focus was on the next frontier—Agentic AI—where multiple LLMs coordinate to handle complex, multi-step tasks. Examples include AI-led infrastructure maintenance and enterprise-wide process orchestration. Agentic systems go beyond assistant-style tools, acting as coordinated agents with memory, decision-making capacity, and process autonomy.
Security & Sovereignty: The presentation emphasized data sovereignty and model training transparency. Hosting LLMs on-premises—such as in Swiss-based data centers for banking clients—ensures regulatory compliance and limits exposure to data jurisdiction risks.
The EU AI Act: With compliance obligations mounting, organizations are urged to implement centralized governance structures such as AI Councils or AI Boards. Transparency, traceability, and reproducibility of model decisions—especially in regulated environments like credit or trading—are non-negotiable.
Technology Infrastructure: Success in AI isn’t just about the model—it requires robust infrastructure. Only 5% of most AI projects involve model development; the rest is “glue and data.” A sound enterprise architecture is critical for scalability, security, and compliance.

Conclusion:
GenAI and Agentic AI represent a structural shift—what the speaker called a “rising tide that lifts all boats.” While the tools are more accessible than ever, true enterprise value depends on integrating these technologies within robust governance frameworks. The message is clear: the era of experimentation is ending, and the time for compliant, scalable, and strategic deployment has arrived. DXC helps clients across the financial services industry to bridge the gap between the enormous potential of Agentic AI and realizing commercial value from the technology.

Sources:

Link to the full presentation video: https://www.youtube.com/watch?v=M05ARutPKkA