AWS re:Invent 2025 - The next frontier: Building the agentic future of Financial Services (INV209)

Summary of AWS re:Invent 2025 - The next frontier: Building the agentic future of Financial Services (INV209)

The Evolving Landscape of Financial Services

• The financial services industry has undergone significant transformation over the past 13 years, driven by the adoption of AWS cloud services and the evolving use cases of customers.
• Financial institutions have been proactive in embracing change, moving their data to AWS and setting up processes to securely leverage machine learning and generative AI.
• This has enabled them to tackle long-standing challenges and transform their operations with agentic capabilities.

The Rise of Agentic Systems in Financial Services

• Agentic systems are increasingly being entrusted with critical responsibilities, interacting directly with customers, handling sensitive data, and executing complex tasks.
• Examples from the presentation include:
  • Moody's deploying multi-agent systems to process data and conduct specialized research and analysis.
  • Ripple transforming security operations by building a multi-agent system to analyze logs and enable guided investigations.
  • Commonwealth Bank of Australia using agents to automatically analyze legacy code, perform security assessments, and handle network flow implementation.

The Vision of Agentic Commerce

• Agentic systems will become more integrated into our daily lives, taking on tasks such as:
  • Searching for and evaluating properties based on personal preferences and financial resources.
  • Analyzing financial situations, evaluating current holdings, and making recommendations to optimize investments and savings.
  • Streamlining insurance underwriting by automatically extracting data and performing risk assessments.
• This autonomous agent-to-agent future requires trust, which will be facilitated by standards and protocols that enable secure, compliant, and data-protective operations.

Building the Foundation for Agentic AI

• Amazon Bedrock, with its expanding selection of foundation models and new features like policy and evaluations, provides a platform for deploying and scaling agentic AI systems securely.

Allianz's Approach to Agentic AI Governance

• Allianz has developed a multi-agent framework that leverages Amazon Bedrock and other AWS services to streamline risk assessment and claims processing.
• Key aspects of their approach include:
  • Defining a hierarchy of agents (utility, business, and planner agents) to enable reusability and scalability.
  • Implementing an "agentic architecture framework" with core components, DevOps tools, and AI models to manage the deployment and governance of agents.
  • Showcasing a use case where agents reduced the claims processing time by 80% for a power outage-related incident.

Coinbase's Vision for Agentic Commerce and Payments

• Coinbase is developing an open standard called X42 to enable internet-native payments that can be seamlessly integrated into agentic systems.
• X42 leverages stable coins and blockchain technology to provide instant, low-cost, and secure payment capabilities that can be programmatically accessed by agents.
• This allows agents to dynamically discover and pay for services, content, and APIs, unlocking new monetization models and improving the overall utility of agentic systems.
• Coinbase's Agent Kit, combined with AWS services like Bedrock, provides a comprehensive solution for building agentic systems with integrated payment capabilities.

Key Takeaways

• Financial institutions are embracing agentic systems to transform their operations and better serve customers.
• Secure and scalable deployment of agentic AI requires thoughtful governance and architectural frameworks.
• Open standards like X42 are emerging to enable internet-native payments and unlock the full potential of agentic commerce.
• The collaboration between financial services providers and cloud platforms like AWS is crucial in shaping the agentic future of the industry.