Summary
BFSI operations frequently encounter significant challenges in efficiently accessing and synthesizing vast, disparate data, leading to delays in critical decision-making for tasks like credit card disputes or claims processing. ServiceNow Now Assist directly addresses this by leveraging generative AI to rapidly provide context-aware information, summarizing complex case histories and customer data directly within workflows. This capability moves beyond simple chatbots, empowering employees with precise, timely insights, thereby enhancing efficiency and decision-making while maintaining crucial compliance and human accountability. Ranosys, with its deep expertise in ServiceNow and AI & Analytics, recognizes this as a pivotal opportunity for digital transformation.
Now Assist demonstrates significant practical impact across banking and insurance. It streamlines customer service by delivering comprehensive pre-call context, accelerates card dispute resolution through automated summarization of transaction details, and enhances loan servicing by drafting responses based on integrated customer histories. For insurance, it notably fast-tracks claims processing and policy servicing. This translates to reduced average handling times, improved first-contact resolution, and greater operational efficiency, enabling employees to focus on expert judgment rather than time-consuming information retrieval. Ranosys's experience in Digital Transformation equips them to identify and implement these high-impact use cases effectively.
Successful Generative AI adoption in BFSI demands robust governance, data security, and clear measurement. Organizations must establish baselines and track key metrics like handling time and AI content acceptance, ensuring quality alongside efficiency. Ranosys, a trusted ServiceNow partner with deep expertise in Digital Transformation and AI & Analytics, guides clients in assessing measurable value, designing tailored governance, and configuring Now Assist for Financial Services Operations. This ensures compliant, effective implementation, optimizing operations and delivering tangible business outcomes while mitigating risks.
It is 9:15 on a Tuesday morning, and a credit card dispute is sitting on an analyst’s screen with a regulatory deadline approaching. Two agents have already added comments, the merchant has responded, and the supporting documents are attached. The actual decision may take two minutes, but getting all the information needed to make that decision can take twenty. When an analyst is handling hundreds of similar cases, those extra minutes can quickly add up and affect resolution timelines.
This is a common challenge for teams working in banking and insurance. The information they need is usually already available across customer profiles, transaction histories, policies, claims, previous interactions, knowledge articles, and case notes. The real challenge is finding and understanding the relevant information quickly enough to act. This is one of the core problems that AI in financial services is being deployed to address.
This is where ServiceNow Now Assist can make a difference across BFSI operations. In this blog, we will look at some practical AI use cases in financial services across banking and insurance, including customer service, disputes, claims, and employee service. We will also cover the implementation considerations, governance, and business impact that organisations need to think about when introducing AI in financial services workflows. At Ranosys, our work with enterprise technology and ServiceNow gives us a practical view of where these capabilities can add value and where human judgement remains important.
Why Generative AI in Banking and Insurance Is More Than a Chatbot
Generative AI in banking and insurance needs to work within a very different set of expectations. Customers want quick answers, employees need the right information at the right time and financial institutions still need to maintain security, compliance, auditability, and human accountability.
This is what makes AI adoption in BFSI different from simply adding a chatbot to an application. Any AI capability needs to work within existing processes and provide the right level of control, especially when customer and financial information is involved.
Customer expectations are changing as well. McKinsey’s 2026 Global Banking Annual Review highlights the rapid adoption of generative AI by consumers for increasingly complex financial tasks. This is putting additional pressure on banks to improve how they deliver digital experiences and support their customers.
For banks and insurers, the opportunity is therefore broader than answering questions through a chatbot. It is about bringing AI into the workflows where employees already work, while keeping governance and human judgement at the centre.
Banking Use Cases for ServiceNow Now Assist
How Now Assist Improves Banking Customer Service
Consider a customer calling the bank about a declined transaction. Generative AI customer service tools like Now Assist can bring the relevant case and customer information into context before the representative even starts the call. Instead of moving between different screens and records to build a picture of the issue, the representative can review a summary of previous interactions, open cases, and relevant knowledge, then focus on resolving the request.
Now Assist can bring the relevant case and customer information into context, helping the representative understand the issue without going through the entire history manually. The representative can review the information, confirm the details, and then focus on resolving the customer’s request.
Instead of spending the first few minutes searching for information, the representative can start the conversation with a clear understanding of what has already happened.
For a single interaction, this may save only a few minutes. Across a high-volume contact centre, those minutes can add up and help teams handle more requests while giving customers a smoother experience.
AI-Assisted Card Dispute Resolution with Now Assist
Card disputes are a good example because most of the information required to process a dispute is already available in the system.
A single dispute may contain:
- Transaction details
- Customer explanation
- Transaction date
- Supporting documents
- Previous interactions
- Merchant response
- Case history
Consider the difference in the analyst’s day:
| Dimension | SiteGenesis | SFRA |
| Architecture | Monolithic, pipeline-based | Modular, controller-based |
| Customization Model | Override or replace core code | Extend functionality through cartridge layering |
| Design Approach | Desktop-first with responsive adaptations | Mobile-first with responsive design by default |
| Template Language | ISML with legacy patterns | ISML with modern, simplified patterns |
| Salesforce Investment | Primarily security and maintenance updates | Active development and ongoing feature enhancements |
| SCAPI and Headless Support | Limited | Built for SCAPI and modern headless implementations |
| LINK Cartridge Access | SG-specific access | Broader catalog access |
| Testing Support | Limited testing capabilities | Supports unit, integration, and functional testing |
| User Experience Optimization | More constrained by its monolithic architecture | Modular architecture enables faster iteration and optimization |
The analyst remains responsible for the decision. The value of AI is in reducing unnecessary reading and documentation, not removing accountability. This distinction is especially important when introducing AI into regulated financial processes.
Loan Servicing Automation: How Now Assist Supports Hardship Processing
A servicing officer handling a hardship request needs to review the payment history, current arrangement, previous interactions, and applicable procedures before responding to the customer. This can involve checking several records before the officer has enough context to respond.
With Now Assist, the officer can start with a summary of the relevant customer and case information and review a draft response based on that context. The officer then checks the details, makes the decision, and remains responsible for the final response.
Instead of spending time going through the customer’s history and writing the response from scratch, the officer can spend that time reviewing the information, making the right decision, and speaking with the customer when needed.
Fraud Investigation and Financial Crime: AI Summarization with Governance
Fraud investigators work with alerts, case notes, previous investigations, customer activity and related cases. One of the biggest challenges is finding the important information within that volume. Summarisation can help investigators understand a case history more quickly and bring related information into view. But this is also an area where governance becomes extremely important.
An AI-generated summary should support an investigation. It should not automatically become the fraud determination. That boundary should be considered as part of the solution architecture from the beginning.
AI-Powered Employee Service and Knowledge Management for Banks
Banks and insurance companies also generate a large amount of internal service demand. Branch employees may need answers about internal policies. New employees may need application access. Operations teams may need to find a procedure or understand how to complete a particular process.
When AI-powered search can answer these questions using trusted enterprise knowledge, some requests may never need to become tickets. This is where ServiceNow knowledge management capabilities within Now Assist make a practical difference: employees find accurate answers through search rather than raising a request, and the service desk handles fewer routine queries.
ServiceNow has reported that its own implementation of Now Assist in Search contributed to a 14% increase in employee deflection rate.
Insurance Use Cases for ServiceNow Now Assist
Generative AI in insurance operations follows the same principle as banking: the information needed to process a claim or service a policy is usually already in the system. The challenge is getting it into context quickly enough to be useful. The following sections cover where Now Assist has the clearest applications across claims and policy servicing.
Generative AI for Claims Processing: Faster Adjuster Workflows
AI in claims processing most commonly addresses the same bottleneck seen in banking disputes: the time spent reading across records before the actual assessment can begin. An insurance adjuster opening a claim may have to review customer and policy information, claim history, adjuster notes, documents, and previous interactions before they can start making progress.
A generated summary can provide that starting point and bring relevant policy information and previous claims into context. The adjuster still reviews the file, validates the policy details, and makes the decision.
What changes is the time required to understand the case? It can also help create a more consistent experience when similar claims are handled by different adjusters or teams.
Policy Servicing Automation with AI and Workflow Integration
Address changes, nominee updates, renewal questions, document requests, and claim-status questions are usually repetitive rather than highly complex. However, they still consume significant employee time. This is where AI and workflow automation can work together.
A typical conversational AI banking and insurance journey using Now Assist and Virtual Agent could look like this: the customer asks a question through a self-service channel, the ServiceNow Virtual Agent identifies the intent, ServiceNow retrieves the relevant knowledge and policy information, AI generates a response in the customer’s language, the workflow performs the required action such as creating an endorsement request, and a human handles cases that require judgement. ServiceNow Virtual Agent and Now Assist are separate capabilities that can be configured and used together as part of the same customer journey.
AI Governance and Data Security for BFSI: What ServiceNow Now Assist Addresses
Generative AI in BFSI needs to be introduced with the same level of care as any other technology that handles customer and financial information. The architecture needs to address questions around:
- Data access
- Identity and roles
- Sensitive information
- Model selection
- Prompt governance
- Auditability
- Human oversight
- AI risk management
Organisations also need to consider the regulatory requirements that apply to their business and region. For example, financial institutions in Singapore need to consider MAS guidelines and requirements, while banks in the US need to consider frameworks and guidance from bodies such as the Federal Financial Institutions Examination Council (FFIEC). The specific requirements will depend on the organisation, use case, and location, so these should be reviewed as part of the solution design.
Knowledge quality is another important consideration. If the underlying knowledge base contains outdated, incomplete or incorrect information, AI can produce responses based on that content. Keeping knowledge articles accurate, current and properly governed should therefore be part of the implementation from the beginning.
Running ServiceNow generative AI capabilities within the same platform where the workflows, records and audit information already exist can simplify part of this discussion. However, organisations still need to validate their specific data, security, privacy, model and regulatory requirements. Model availability and capabilities can also vary by ServiceNow release, region, and subscription, so these details should be confirmed during solution design.
The industry is already moving in this direction. In January 2026, ServiceNow and Fiserv announced an expanded commitment to scale Now Assist for Financial Services Operations and IT Service Management across environments supporting Fiserv’s financial-services clients. In July 2026, ServiceNow also announced a USD 40 million investment in BUSINESSNEXT to accelerate AI-native financial-services solutions across Asia Pacific.
Measuring the Business Impact of Now Assist in Financial Services
A successful demonstration is not the same as a successful implementation. This is where many AI initiatives can lose momentum.
Before enabling an AI capability, establish a baseline.
Customer service
- Average handling time
- First-contact resolution
- Case volume
- Reopen rate
- Customer satisfaction
Operations
- Cases processed per employee
- Average processing time
- Manual documentation time
- Escalation rate
Self-service
- Deflection rate
- Virtual Agent containment
- Search success rate
- Human handoff rate
AI quality
- AI-generated content acceptance rate
- Correction rate
- User feedback
- Reopened cases
The last group is particularly important. If cases are being closed faster but reopened cases are increasing, the organisation may have improved speed at the expense of quality.
In that situation, the problem is not necessarily the technology. It may indicate that the review process, training, or governance needs to be improved.
What Financial Services Leaders Should Know Before Implementing Now Assist
Generative AI in BFSI has moved beyond experimentation and is finding practical applications in customer service, financial operations, claims, knowledge management and employee service. Successful adoption is not simply about switching on an AI feature. It requires a combination of:
Trusted data + Enterprise knowledge + Generative AI + Workflow automation + Governance + Human judgement
ServiceNow Now Assist provides a practical foundation for this approach, particularly through Now Assist for Financial Services Operations. For a leadership team, the better question is not: “What can AI automate?” Instead, ask: “Where are our people spending time reading, searching, and documenting when they should be deciding, serving, and solving?”
That is where the opportunity is. And it is something that can be measured from the first pilot.
The Tuesday morning dispute will still arrive. The difference is that the analyst no longer needs to spend the first twenty minutes reconstructing the story before starting to solve it.
How Can Ranosys Help?
Ranosys is a ServiceNow partner with implementation experience across banking, insurance, and capital markets operations. We work with BFSI technology and operations leaders to assess where Now Assist creates measurable value in their specific workflows, design governance frameworks that meet the regulatory requirements of their region, and configure Now Assist for Financial Services Operations alongside Virtual Agent, AI Search, and existing ITSM or CSM setups.
Our approach starts with a baseline. Before any capability goes live, we establish the metrics that will show what has changed: handling time, deflection rate, AI content acceptance, and reopen rates. This gives leadership a clear picture of what the implementation is delivering from the first pilot onward.
If your team is evaluating ServiceNow generative AI for banking or insurance operations and wants a practical view of where to start and what to govern, we are happy to walk through it with you.
Contact our ServiceNow team to discuss your current workflows, review what Now Assist for Financial Services Operations covers in your environment, and identify where AI in financial services can reduce the overhead your teams are carrying today.
Rahul Swami
Project Lead
Rahul Swami is a Project Lead at Ranosys with extensive experience delivering technology solutions for the BFSI sector. He has worked closely with financial institutions, gaining a strong understanding of their business processes, operational needs, and technology challenges. His expertise spans low-code development and solution architecture, with hands-on experience in ServiceNow and OutSystems, and a focus on using generative AI to address evolving BFSI needs. Connect with him on LinkedIn.
