View this page in the latest version of Appian. AI in Appian Insurance Solutions Share Share via LinkedIn Reddit Email Copy Link Print On This Page Why AI in insurance Claims and underwriting work is document-heavy and research-heavy. Adjusters read policy documents to confirm coverage and comb through long comment threads to understand claim history. Underwriters read submission packets to assess risk before they can quote. Life underwriters review medical and financial evidence attached to each case. In all three workflows, professionals spend significant time locating facts before they can act on them. Appian's insurance solutions embed artificial intelligence directly into these workflows so that AI works alongside claims and underwriting professionals rather than replacing their judgment. Every AI feature in the insurance solutions is designed around a simple principle: AI suggests, humans decide. Benefits at a glance The following table summarizes the core benefits of AI across the Appian insurance solutions: Benefit How AI delivers it Example Faster document review Generates summaries of uploaded PDF documents so users grasp the content without reading it end to end. AI Claim Document Summary and Case Document Summary produce summaries as documents are attached. Faster policy research Lets adjusters chat with lengthy policy documents to locate the terms, exclusions, and limits that matter, instead of reading each document end to end. AI Policy Document Assistant answers questions about a policy document from the policy interface in Connected Claims. Answers in context Lets users ask questions of a record and its related data and documents in natural language. AI Chat with Claim and Case Record Chat answer questions from the claim or case record itself. Faster coverage decisions Analyzes policy terms against loss details and returns a suggested coverage status. AI Coverage Analyzer and Claim Coverage Verification AI surface coverage findings on new claims. Better work routing Evaluates incoming work for severity, complexity, and urgency and recommends an assignment. Claim Triage Agent recommends the handling team for a newly reported claim. Reduced manual data entry Extracts structured data from inbound documents and correspondence. ACORD Document Extraction populates submission fields from ACORD forms received by email, and Create Claim via Email extracts claim data from adjuster inboxes. Quick thread context Condenses long comment and message histories into a short readable summary. AI Claim Comment Summary and OpenAI Messaging Summary summarize correspondence history. Appian's AI-first approach Appian takes a Private AI approach. The Private AI approach means: Your data stays yours. Customer data is never shared outside the Appian Cloud compliance boundary and is never used to train Appian's models. Your models are never shared with other customers. No specialized expertise required. AI features are configured through low-code interfaces, using the same design patterns insurance business users and developers already use in Appian. Purpose-driven AI. Each AI feature is scoped to a specific task, such as summarizing a document, extracting submission data, or evaluating coverage, rather than offering open-ended, unconstrained generation. By default, AI features in the insurance solutions use GPT 5.4 through Appian's Private AI service. GPT 5.4 powers generative capabilities, such as drafting, summarizing, and answering questions, across Connected Claims, Connected Underwriting P&C, and Connected Underwriting Life Workbench without requiring additional model configuration. Appian's AI capabilities use both discriminative models, which classify and extract, and generative models, which draft and summarize. For a deeper introduction to how these models work, see What is Artificial Intelligence? in the Appian platform documentation. Bring your own model While the default GPT 5.4 model covers most use cases out of the box, Appian also supports bringing your own AI model. Carriers with specific model requirements, such as particular authorization boundaries, data residency constraints, or preferred providers, can configure alternative models at the platform level. Model flexibility also benefits organizations that want to evaluate different models for cost, performance, or capability reasons. Administrators and developers can select models at multiple levels: Enable and configure models for your environment. Use the Admin Console to activate AI models and manage inference profiles that control which models are available to your applications. See AI Services and Enable AI models and inference profiles. Choose a model in an AI Skill. When building AI Skills with Prompt Builder or Generative AI Skills, developers can select a specific model for each skill. See Prompt Builder: Select a model and Generative AI Skills: How to select a model. Choose a model for an AI Agent. AI Agents can be configured to use a specific model and provider during creation. See Create and configure AI agent: Select a model and provider and AI Agent Reference: Model and provider selection. Some insurance AI modules can also be pointed at Azure OpenAI instead of Appian's AI services. For an example, see the Appian Connected Underwriting P&C OpenAI Module. Security and compliance AI features in the insurance solutions inherit the platform's security posture. Appian does not build a separate AI security model for solutions, so the protections that apply to your claims and underwriting data apply to AI processing of that data. Private AI principles Appian's Private AI model ensures that: Customer data is never shared outside the Appian Cloud compliance boundary, and is never used to train Appian's models or shared with other customers. Your data always remains in your control. Content safety is enforced at the platform level through AI Guardrails, which evaluate AI interactions for policy violations. Every AI-driven action is logged and auditable within the platform's process monitoring. For full details on Appian's AI governance commitments, see Responsible AI. For the current list of certifications and Appian's compliance posture, see Security and Compliance. Customer control Administrators retain full control over AI features in their environment: Modular adoption: Each AI capability is an add-on module you install and set up independently. Your base claims or underwriting application keeps working whether or not you adopt AI. Permission-aware output: AI features operate with the initiating user's permissions, so users only see AI output drawn from records and documents they can already access. Guardrails: Configure constraints on AI behavior to align with carrier policies. See AI Guardrails. Human-in-the-loop: AI outputs are presented as suggestions. Users review the underlying document or record and decide before any action is taken. Audit trail: AI interactions are recorded in process history for compliance review. AI modules in insurance solutions Unlike solution families that share a common set of AI modules, each Appian insurance solution ships its own AI add-on modules, scoped to the workflow it supports. This means you adopt AI where it removes work from a specific claims or underwriting team, and skip it where it doesn't. Each solution publishes a landing page that lists its AI modules along with installation and setup instructions: Solution AI module documentation Appian Connected Claims AI Add-on Modules Appian Connected Underwriting P&C AI Modules Appian Connected Underwriting Life Workbench AI Modules Because the modules are independent, an AI module release does not force a base application upgrade, and you can pilot one module before adopting others. Built-in AI features by insurance solution The insurance solutions embed AI capabilities directly into each claims and underwriting workflow. The following sections summarize the AI features in each solution. For detailed configuration and usage instructions, follow the links to solution-specific documentation. Appian Connected Claims Appian Connected Claims offers seven AI add-on modules that support the claim lifecycle from first notice of loss through settlement: AI Chat with Claim: Ask questions about a claim in natural language and receive answers drawn from that claim's records and related data, without navigating between views. AI Claim Comment Summary: Condenses long claim comment threads into a short summary so adjusters and supervisors can pick up claim history quickly. AI Claim Document Summary: Generates summaries of PDF documents uploaded to a claim, so users understand the content without opening each file. AI Coverage Analyzer: Analyzes new claims against policy coverage terms and returns a suggested coverage status for adjuster review. AI Policy Document Assistant: Chat with policy documents to locate terms, exclusions, and limits relevant to the claim at hand. Claim Coverage Verification AI: Verifies claim details against the policy to surface coverage findings and support the coverage decision. Claim Triage Agent: Evaluates incoming claims and recommends routing and team assignment based on claim characteristics. For overviews, installation, and setup for each module, see AI Add-on Modules in the Connected Claims documentation, and the individual module pages: AI Chat with Claim, AI Claim Comment Summary, AI Claim Document Summary, AI Coverage Analyzer, AI Policy Document Assistant, Claim Coverage Verification AI, and Claim Triage Agent. Connected Claims also applies AI outside of the AI add-on modules: AI-assisted claim creation: The Create Claim via Email module ingests mail from designated claims inboxes into a triage queue, uses AI to extract claim data fields from the email content, and lets adjusters create a claim directly from the email with those fields already populated. This capability ships with the Create Claim via Email module rather than with an AI add-on module. For intake configuration and the full list of capabilities, see Create Claim via Email in the Connected Claims documentation. Note: Connected Claims only comes out-of-the-box with the advanced and premium capability tiers. Check each module's documentation for its specific tier and licensing requirements. Appian Connected Underwriting P&C Appian Connected Underwriting P&C applies AI to submission intake, document review, and account analysis: Submission AI: Ask questions about a submission in natural language from the Submission Summary view, and generate a summary covering its key risks and the substance of its notes. This module builds on Appian AI. AI Submission Document Assistant: Chat with submission documents to find risk details, schedules, and loss history without reading each file. OpenAI Messaging Summary: Summarizes message threads on a submission so underwriters get quick context on correspondence history. This module connects to Azure OpenAI. ACORD Document Classification: Identifies the ACORD form type (125, 127, or 140) for documents polled from email before extraction runs, using a document classification AI skill. Documents falling below a configurable confidence threshold route to a classification exception queue for human review. This capability is part of the base application and does not require an add-on module. ACORD Document Extraction: Extracts data from classified ACORD forms to populate a submission, with a reconciliation task for underwriter validation. Extraction is supported for up to 125,000 documents per month. This capability is part of the base application and does not require an add-on module. AI financial performance summary: Generates a natural language summary of an account's financial performance from the underwriting data already in the solution. This capability is part of the base application and does not require an add-on module. For module overviews, installation, and setup, see AI Modules in the Connected Underwriting P&C documentation, and the individual pages: Submission AI, AI Submission Document Assistant, and OpenAI Module. For the financial performance summary, see Managing Submissions. For document classification and extraction configuration, see Set Up the Solution. Appian Connected Underwriting Life Workbench Appian Connected Underwriting Life Workbench applies AI to the evidence review that drives life underwriting decisions: Case Document Summary: Generates summaries of PDF documents uploaded to a case, using the Document Summarizer AI Skill, so underwriters understand medical and financial evidence faster. Case Record Chat: Ask questions about a case in natural language and receive answers drawn from that case record, built on the Records Chat Component and Appian AI. For module overviews, installation, and setup, see AI Modules in the Connected Underwriting Life Workbench documentation, and the individual pages: Case Document Summary and Case Record Chat. Set up AI in insurance solutions To begin using AI features in your insurance solutions: Verify that your environment meets the prerequisites. AI features require an Appian Cloud environment with AI services enabled for your site. See AI Services. Contact your Appian account team to confirm AI module licensing and capability tier requirements for the modules you plan to adopt. Install the AI module you need from the solution's AI documentation. Each module has its own installation and setup instructions, and each is independent of the base application. Complete the module's setup steps to configure the AI service connection, prompts, and any feature-level settings, then verify the feature against sample claims, submissions, or cases. Tip: Start with a single AI module in a non-production environment to familiarize your team with the human-in-the-loop review workflow before rolling out broadly. Additional AI and platform documentation The following Appian platform documentation provides additional context on AI capabilities, terminology, and troubleshooting: What is Artificial Intelligence?. Introduction to AI concepts and how Appian applies them. Private AI. Appian's approach to protecting your data during model creation, training, and use. AI Services. How AI services are provisioned and which models are available in your environment. AI Guardrails. How to configure constraints and safety boundaries for AI features. Responsible AI. Appian's commitments to ethical, transparent, and accountable AI. Security and Compliance. Full list of authorizations, certifications, and security controls. AI Glossary. Definitions of AI terminology used across Appian documentation. AI Troubleshooting. Common issues and resolution steps for AI features. Feedback Was this page helpful? SHARE FEEDBACK Loading...