Leveraging ChatGPT In Legal Practice: Best Practices And Ethical Considerations

You can think of ChatGPT and tools like it as a highly intelligent assistant who sometimes sprinkles inaccuracies into its otherwise very helpful responses.

artificial-intelligence-3382507__340By now, you’ve likely heard about generative AI and ChatGPT and are wondering if it lives up to the hype. As I discuss in an earlier post, I believe it does and that legal professionals need to learn about it and understand how and when to use it.

That being said, a key part of learning about this emerging technology includes understanding its limitations and risks. Over time, as the technology improves and is incorporated into legal technology products specifically designed with the needs and compliance requirements of legal professionals in mind, the risks will be reduced.

But for now, there are a number of issues to keep in mind should you choose to use tools like ChatGPT in your law firm.

What Is Generative AI?

Generative AI tools are essentially artificial intelligence chatbots, trained on broad datasets, that are able to respond in a conversational and informative manner to many types of inquiries.

You can think of ChatGPT and tools like it as a highly intelligent assistant who sometimes sprinkles bald-faced lies into its otherwise very helpful responses. These factual inaccuracies are often referred to as “hallucinations.”

The most common commercially available examples of this type of technology include ChatGPT, Google Bard, and Microsoft Bing Chat. Additionally, legal software platforms are increasingly incorporating GPT-4 powered generative AI tools into their systems as well.

Understanding The Risks, Benefits, And Best Practices

These AI tools have tremendous potential, and their thoughtful and strategic adoption into law firms can offer significant advantages. There are many ways that lawyers can use generative AI to reduce the pain of time-consuming (and sometimes mundane) tasks.

Some examples include:

  • Drafting documents, forms, and templates
  • Summarizing articles, correspondence, laws, regulations, and cases
  • Contract review and analysis
  • Litigation preparation, including sample questions for voir dire or cross-examination
  • Brainstorming ideas.

Now that you understand the basics of generative AI and the benefits it can offer legal professionals, let’s move on to risk reduction. A number of steps can decrease risk and increase accuracy when implementing this technology in your firm.

Legal Ethics Issues

First and foremost, always keep client confidentiality top of mind. It’s imperative that you fully understand how any data queries are handled and who has access to them. If you are unable to adjust the product’s privacy settings to reduce data sharing for product improvement purposes or otherwise, you’ll need to refrain from including any confidential client information when submitting queries to the chatbot.

You can learn more about other potential legal ethics you may encounter issues in this post.

The Art Of The Prompt

You’ll also need carefully craft prompts so that you’ll obtain the desired response. The key elements to a successful query are:

  • Specificity: Provide clear and detailed prompts.
  • Context: Offer enough background information for the topic.
  • Instruction: Explicitly request the type of response you want.
  • Proper Formatting: Use correct grammar and structure in your prompts.
  • Understanding AI’s Limitations: Keep in mind the AI’s limitations, including knowledge and date cutoffs.

One way to ensure that the input provided is sufficient is to rely on a tool like Prompt Perfect (either as a standalone tool or plugin to ChatGPT), which will transform your basic request for output into a detailed and contextual prompt.

Don’t Fall For Hallucinations

Lastly, because available generative AI tools continue to provide responses that include inaccuracies and flat-out falsehoods, it’s critical that you have adequate foundational knowledge on all topics you’ll be asking about. Then you’ll need to carefully review all answers provided so that you can identify any errors in the output.

At this early stage, the importance of knowledgeable human review of output cannot be overstated. However, as generative AI technology advances, less oversight will be required, and the overall quality of results will exponentially improve.

The Future Is Bright

The bottom line: Don’t let concerns about potential risks prevent you from taking advantage of the many benefits generative AI offers. Tools like ChatGPT present compelling opportunities for law firms, and by employing best practices — such as crafting precise, context-rich queries — and fully comprehending AI’s current limitations, legal professionals can leverage these technologies effectively and responsibly, to the benefit of their law practices and clients.


Nicole Black is a Rochester, New York attorney and Director of Business and Community Relations at MyCase, web-based law practice management software. She’s been blogging since 2005, has written a weekly column for the Daily Record since 2007, is the author of Cloud Computing for Lawyers, co-authors Social Media for Lawyers: the Next Frontier, and co-authors Criminal Law in New York. She’s easily distracted by the potential of bright and shiny tech gadgets, along with good food and wine. You can follow her on Twitter at @nikiblack and she can be reached at niki.black@mycase.com.

Sponsored

CRM Banner