Top AI Product

Potential Uses of AI Agents in Banking and Insurance

AI agents are intelligent systems designed to process and analyze vast amounts of financial data, assisting in decision-making and predicting outcomes. Their potential uses span several key areas:

• Customer Service in Banking and Insurance:
  AI agents can serve as virtual assistants, offering 24/7 support for customers. They can handle inquiries, assist with transactions, and provide information about products and services. For example, in banking, they can answer questions about account balances or loan terms, while in insurance, they can help with policy inquiries or claims processing. Research suggests 89% of service professionals at financial institutions report increased customer expectations for personalized experiences, and AI agents can meet these demands by reducing wait times and improving accessibility. Continuous monitoring is another area, particularly for fraud detection, where AI agents can send alerts for suspicious activities. Future advancements in natural language processing (NLP) and machine learning could make these assistants more sophisticated, enabling early intervention by analyzing customer data patterns for potential issues.
• Risk Management:
  AI agents are poised to enhance risk assessment by analyzing large datasets, including transaction histories, credit scores, and market trends, to detect fraud and ensure compliance. In banking, they can identify suspicious transactions in real-time, flagging them for further investigation, often before significant harm occurs. In insurance, AI agents can assess risks for underwriting, determining premiums based on individual risk profiles, and detect fraudulent claims by identifying anomalies. A report from KPMG indicates 68% of banking executives plan to use generative AI for compliance and risk, highlighting the growing reliance on AI for these tasks. Predictive analytics could further enable proactive risk mitigation, shifting from a “detect and repair” to a “predict and prevent” framework, as noted by the NAIC .
• Personalized Financial Advice:
  AI agents can tailor recommendations based on individual financial profiles, enhancing customer engagement. In banking, they can offer investment advice or suggest savings strategies, while in insurance, they can recommend personalized policy options. This level of personalization, delivered consistently at scale, can enhance customer satisfaction and loyalty, with 64% of organizations expecting AI agents to relieve human workers of repetitive tasks, allowing focus on value-added functions like customer experience . Future improvements are expected to leverage behavioral data for more nuanced advice, though ethical concerns around data privacy and bias remain.
• Automating Routine Tasks:
  AI agents can streamline administrative tasks, reducing operational costs. In banking, they can automate account opening, transaction processing, and compliance checks, freeing employees for strategic work. In insurance, they can handle claims processing, policy issuance, and document verification, improving efficiency. For instance, agentic AI can reduce the time taken for loan documentation and transaction checks, with operations like data entry being tended to by software, as noted in a recent analysis . This automation is expected to drive productivity, with banks and insurers potentially saving billions through reduced manual labor, though challenges like job displacement are debated.
• Emerging Trends:
  Several trends highlight the future potential of AI agents. Wearable devices and IoT integration are evolving, particularly in insurance, with usage-based insurance (UBI) models like pay-by-mile car insurance or pay-by-stay home-sharing insurance becoming norms . Predictive analytics is shifting finance towards proactive strategies by identifying high-risk individuals or predicting market shifts. The convergence of these trends points towards a personalized, efficient, and effective financial ecosystem, though challenges like data privacy and ethical AI use persist. A notable quote from Jamie Dimon of JP Morgan underscores this: “AI will likely augment virtually every job,” highlighting its transformative potential

Additionally, while not as prominently featured, AI agents could extend to mental health support in financial stress management or elderly care in insurance, assisting with policy reminders and monitoring. Public health applications, such as epidemic prediction for insurance, are also potential areas, though less developed in current discussions. These areas, particularly mental health and elderly care, require careful consideration of ethical frameworks to ensure fairness and trust.

Existing Products in Banking and Insurance

Several AI agent products are already deployed in finance, demonstrating practical applications and tangible benefits:

• Klarna’s AI Assistant:
  Launched in February 2024, Klarna’s AI assistant, powered by OpenAI, handles customer service chats, doing the equivalent work of 700 full-time agents. It has had 2.3 million conversations, representing two-thirds of Klarna’s customer service chats, and is on par with human agents in customer satisfaction, with a 25% drop in repeat inquiries. Customers now resolve errands in less than 2 minutes compared to 11 minutes previously, available in 23 markets, 24/7, and communicates in over 35 languages. It’s estimated to drive a $40 million USD profit improvement to Klarna in 2024, enhancing communication with local immigrant and expat communities (Klarna AI Assistant). An unexpected detail is its language support, significantly improving accessibility for diverse populations.
• Bank of America’s Erica:
  Launched in 2018, Erica is a virtual financial assistant with over 2 billion interactions, helping 42 million clients. It provides personalized financial advice, assists with transactions, and identifies savings opportunities, acting as both a personal concierge and mission control. It took four years to reach 1 billion interactions, with engagement surging to reach a second billion in just 18 months, supported by predictive analytics and language processing. Capabilities have expanded to support individual and corporate clients across Merrill, Benefits OnLine®, and CashPro® platforms, with 333 million engagements in 2023 alone, up 35% year-over-year (Bank of America’s Erica). An unexpected detail is its integration with corporate platforms, extending beyond retail banking.
• Allstate’s ABIE:
  Introduced in 2018, ABIE (Allstate Business Insurance Expert) is an AI-powered tool that assists small business clients with insurance-related questions and locates critical documents via an onscreen avatar. It learns on the go, adding questions and answers to its repertoire, based on structured content and a platform like easyDITA. It enhances customer experience by addressing coverages, incidents covered, and more, improving productivity for agents (Allstate’s ABIE). An unexpected detail is its focus on small businesses, a niche often underserved by AI solutions.
• Cognigy’s AI Agents:
  Positioned as a leader in the Aragon Research Globe™ for AI Agent Platforms, 2025, Cognigy offers conversational AI solutions for insurance, delivering seamless service experiences for customers worldwide, on any channel and in any language. Pre-trained AI agents automate tasks like policy inquiries, claims processing, and customer support, with integration into over 30 voice and digital channels out-of-the-box. An example is Virgin Pulse enhancing member support by integrating a fully trained AI agent with Zendesk LiveChat, simplifying manual processes and lowering customer effort (Cognigy’s AI Agents for Insurance). An unexpected detail is its focus on interoperability, ensuring seamless integration with enterprise systems.
• SalesForce’s Agentforce:
  Rolled out in 2024, Agentforce is a set of tools to create and customize AI agents, with preconfigured agents for financial services, using Salesforce data, Atlas Reasoning Engine, and industry-specific data models. It ensures trust, compliance, and security, automating regulatory compliance tasks. Use cases include wealth management for personalized advice, banking for 360-degree customer views, and insurance for claims management. Statistics show 82% of large companies plan to implement agents by 2027, with 85% of customer service reps at AI-using orgs saying it saves time (SalesForce’s Agentforce). An unexpected detail is its focus on regulatory compliance, addressing a critical need in finance.

Other notable mentions include platforms like H2O.ai, which provide AI solutions for insurance, such as fraud detection and claims management, though not specifically agent-focused. JP Morgan has internal AI assistants for employees, like LLM Suite, but lacks a widely publicized customer-facing AI agent.

Conclusion

AI agents in finance are at an exciting juncture, with current products demonstrating tangible benefits and future potential pointing towards a more personalized, efficient, and proactive financial ecosystem. While challenges like data privacy, job displacement, and ethical AI use persist, the integration of AI with human expertise, as noted by Jamie Dimon, underscores their role as augmentative tools. This analysis provides a foundation for understanding their transformative impact, current as of March 9, 2025.