Artificial intelligence (AI) is becoming an increasingly valuable tool for claims professionals, helping to streamline everyday tasks, analyse information and improve efficiency.
However, the quality of AI-generated responses depends heavily on the instructions provided. These instructions, known as prompts, help AI tools understand what you need, why you need it and how the information should be presented.
This guide outlines the principles of effective prompting, provides practical tips and explores how AI tools can support common tasks in the claims environment.
Why effective prompting matters
A well-structured prompt helps AI tools generate relevant, accurate and useful responses.
Much like briefing a colleague, providing clear instructions and sufficient context helps ensure the output meets your expectations. The more specific you are about the task, intended audience and desired outcome, the more useful the response is likely to be. Effective prompting can also reduce the need for repeated adjustments, saving time and helping claims professionals work more efficiently.
The five golden rules of prompting
An effective prompt typically includes five key elements:
1. Action
Clearly explain what you want the AI tool to do. For example, “Produce an executive claims summary…”
2. Context
Explain the purpose of the task and who the output is intended for. For example, “…for my manager to review…”
3. Sources
Identify the information or documents the AI should use. For example, “…using the attached PDF report…”
4. Constraints
Specify any requirements, limitations or expectations. For example, “…highlight only the most important developments and keep the summary concise…”
5. Format
Describe how you would like the response presented. For example, “…using bullet points that can be read in under two minutes.”
Putting it all together
“Produce an executive claims summary for my manager using the attached PDF report. Highlight the key developments, decisions and outstanding actions. Present the information in concise bullet points that can be read in under two minutes.”
By combining these five elements, you can provide AI tools with clearer direction and increase the likelihood of receiving a useful response.
Practical tips for writing effective prompts
When using AI tools, consider the following approaches to improve the quality of your results:
AI can help extract and summarise relevant information from reports and supporting documents to populate standardised templates.
Example prompt:
“Complete the Market Development Communication template using the information in the attached report. Summarise the key developments relevant to the market for participating insurers who require an overview to support decision-making. Include only information supported by the report and present the output using the existing template structure.”
AI tools with appropriate access to email systems can help identify relevant correspondence, saving time when reviewing large volumes of communications.
Example prompt:
“Search my email inbox for messages received within the last six months that contain the CLA Code ‘QUAN’. This is for an internal review. Present the results in a table showing the sender, recipient, date received, subject line and a brief summary of each email’s relevance.”
AI can support the initial review of claims and loss data, helping identify unusual patterns, outliers or developments that may require further investigation.
Example prompt:
“Review the attached loss data and identify any unusual trends between January 2025 and December 2025. This analysis is for an internal claims review, so focus on anomalies, outliers, sudden changes and emerging patterns. Summarise the findings in a short, plain-English report with key insights presented as bullet points. Highlight any areas that may warrant further investigation.”
AI can help claims professionals prepare clear, concise communications that explain the current position of a claim and any outstanding actions.
Example prompt:
“Using the attached claim notes, draft a concise update for the broker explaining the current position and any outstanding actions. This is a routine claims progress update, so focus on material developments, decisions and next steps. Use clear, professional and plain-English language. Present the information as a short, email-ready summary or four to six bullet points.”
AI can help simplify complex claims correspondence, making information easier for insureds to understand while retaining essential details.
Example prompt:
“Using the attached letter, rewrite the content in plain English for the insured. This is a claims update, so ensure the tone is clear, supportive and easy to understand. Explain the current position, key facts and any required next steps. Retain all policy conditions displayed in italics exactly as written and present the revised letter in a Word document.”
Using AI responsibly
While AI can improve efficiency and support claims-related activities, it should complement rather than replace professional judgement.
All AI-generated outputs should be carefully reviewed and validated before use, particularly where they involve claims decisions, policy interpretation or external communications.
Claims professionals should also follow their organisation’s AI usage policies and information security requirements, ensuring that confidential, personal or commercially sensitive information is handled appropriately and only shared with approved AI tools.
Effective prompting, combined with appropriate human oversight, can help claims professionals make better use of AI while maintaining the accuracy, quality and professional standards expected across the Lloyd’s market.