Pavlé Sabic, Senior Director of Generative AI Solutions and Strategy, speaks with Emerj to give his take on how the emergence of Agentic AI will impact financial services in the technology's early adoption phase, as well as when Agentic AI's influence expands.
The future of AI is expanding beyond single-task automation. As the demand for smarter, enterprise-grade solutions increases, Agentic AI will continue to permeate the regulated financial services industry — an industry facing ongoing pressure to keep pace with a turbulent operating environment without sacrificing quality, compliance, and human expertise.
What can Agentic AI do for finance?
Imagine a team of specialized digital agents working in collaboration to help automate tasks that can take days to complete when executed manually. This intelligent technology is designed to orchestrate complex workflows while interacting with proprietary data, adapting to regulatory guardrails, and delivering consistency across audit-ready outputs.
Tackling challenges with a “human-in-the-loop” approach
Financial institutions are faced with myriad challenges, including fragmented or unreliable data, manual inefficiencies, regulatory pressures, and overburdened analysts. For AI to address these effectively, it must also deliver on transparency and accountability.
As Sabic explains, “The real magic isn’t just in what large language models (LLMs) can do—it’s in their ability to be audited. That’s what makes agentic AI viable for enterprise use.”
Which is why, despite the power of these AI models, human oversight remains at the core of the workflow. Agentic AI can mitigate common institutional challenges by aiding in the automation of documentation and supporting decisions efficiently. Analysts, however, must still define the process, validate outputs, and finalize decisions.
Proprietary data: The foundation for Agentic AI
Moody’s Agentic Solutions are built on decades of proprietary data, including Moody’s Ratings credit research, reports, firmographics, and risk content. This foundation helps enable agents to execute client-specific workflows with precision for regulatory environments.
Implementing a workforce-ready strategy
Successful adoption of agentic AI requires an understanding of the current workflows and processes at an institutional level. This helps ensure alignment and supports enterprise-grade orchestration and workflow redesign.
As noted by Sabic, implementation is more than just deployment. Aligning the technology with core business priorities is how real transformative change is achieved.
Looking ahead
The emergence of Model Context Protocol (MCP) represents a shift from single-task automation to multi-agent orchestration. This means LLMs can use multiple tools to simultaneously complete complex workflows across platforms , subject to human oversight. While adoption is still in the early stages, the transformative potential looks promising.
Embracing the emergence of Agentic AI in finance
To explore Pavle Sabic’s full perspective on Agentic AI, listen to the full interview on the Emerj Business in AI podcast.
For more on how we’re delivering secure, auditable AI at enterprise grade, visit our Agentic solutions page.
Interested in speaking with us? Click on the link below.
Learn more about Moody's Agentic solutions
Learn More
Beyond prompts: Why enterprise AI demands context engineering
Enterprises have begun to discover what the GenAI hype can obscure: large language models are convincing but inconsistent unless fed the right data. That’s why the true differentiator in enterprise grade GenAI isn’t style, but substance — specifically, context engineering.