AI Tools Now Let Lawyers Personalize Work Styles

Lawyers often find themselves repeating the same instructions when using AI tools like Copilot, ChatGPT, and Claude. These tools now offer personalization features to streamline this process, allowing users to establish standing guidance about their work style and preferences. This reduces the need to retype instructions for every new conversation, saving time and enhancing productivity.

Teach Your AI How You Work

Personalization features fall into two categories: standing instructions and memory or persistent context. Standing instructions provide direction on tone, format, and process, while memory retains information like your role, practice area, and recurring audience across conversations. For instance, standing instructions might include directives like “Keep it concise” or “Use plain English,” while memory could store details such as “I frequently work on employment matters for employers with 50 to 500 employees.”

For example, a lawyer might instruct an AI to “Use headings and bullets when they improve readability” or “If important information is missing, ask before making assumptions.” Memory, on the other hand, could store details such as “I am a commercial litigator practicing primarily in Washington state.” This combination ensures the AI understands both how to work and what to remember, reducing repetitive prompts.

AI Personalization in Action

Each AI tool implements personalization differently. Microsoft Copilot offers Custom Instructions and Saved Memories, while ChatGPT provides Custom Instructions and Memory. Claude emphasizes project-scoped memory, keeping context separate for different clients or matters. These features are particularly useful for legal professionals who handle multiple projects or clients simultaneously.

A lawyer using ChatGPT might instruct it to “For short writing requests, provide two alternatives: one direct and one more diplomatic.” Another using Claude could specify, “Write in a clear, restrained style.” These instructions ensure the AI aligns with the lawyer’s preferences while maintaining clarity and professionalism.

To build effective personalizations, start by engaging in a self-interview with the AI’s help. Use a prompt like, “Ask one question at a time about my role, typical assignments, audience, tone, preferred formats, recurring workflow, risk tolerance, how you should handle uncertainty and what I dislike in AI responses.” Review and refine the results, testing them on routine tasks to ensure they align with your needs. This iterative process ensures the AI adapts to your specific workflow.

Read Also: Running Improves Trial Attorney Skills

Personalization serves as a starting point, not a delegation of judgment. It reduces repetition and cleanup, helping your AI tools better understand your work style from the outset. For example, if you frequently use Microsoft 365, you can include instructions tailored to Outlook, Word, Teams, OneDrive, and SharePoint, ensuring recommendations are relevant to your tools.

What to Include in Personalization

When building personalizations, it’s essential to distinguish between what belongs in standing instructions and what should remain in the current prompt or matter workspace. Suitable candidates for personalization include stable tone and formatting preferences, your role, practice area, and typical audience. Recurring processes, rules for handling uncertainty, and preferred terminology are also appropriate. For instance, specifying that you prefer practical recommendations and risk flags over lengthy explanations can streamline outputs.

However, certain elements should not be included in personalization settings. Confidential client names, facts, and strategies, as well as one-time deadlines or short-lived assignments, are better kept in the current prompt. Passwords, access keys, and any information prohibited by firm policy or client terms should never be stored in personalization settings. This ensures security and compliance with information-governance policies.

Creating and Refining Personalizations

To create effective personalizations, consider using the AI tool itself as an interviewer. Begin a chat with a prompt like, “Interview me to create a concise set of personalization instructions for my work.” The AI can ask questions about your role, typical assignments, audience, tone, and preferred formats. After the interview, review the drafted instructions, removing any temporary or sensitive information. Test these instructions on routine tasks and refine them as needed to ensure they align with your work style. This method leverages the AI’s capabilities to help you define your preferences efficiently.

For legal professionals, the focus should be on creating instructions that go beyond mimicking your style. They should encourage the AI to act as a careful assistant, surfacing uncertainty, verifying information, and leaving final judgment to the lawyer. This approach ensures that personalization enhances, rather than replaces, your professional expertise. By tailoring instructions to your specific needs, such as keeping responses under 150 words or using specific terminology like “team member” instead of “employee,” you can ensure the AI aligns with your professional standards.