AI IN THE LEGAL WORLD: IS YOUR INFORMATION STAYING PRIVATE?~3 min read

AI is everywhere, and like many industries, the legal world has been starting to utilize AI in the practice of law. However, there are many considerations that must be accounted for before utilizing AI. Perhaps the most important consideration is your privacy. One of the methods AI tools use to gather data is information that users provide to the tool. This means that when you enter information into ChatGPT, it retains that information for future use. Many AI sources try to assuage privacy concerns on the part of users by trying to give users the impression that they are in control of whether and how their data is stored and shared, but these assurances are often misleading at best.

Of course, because the AI tools need your data to improve their functionality, they are portraying the choice to share your data as a good thing. Open AI even published an article in March of 2026 called “How your data is used to improve model performance.” The article states that Open AI “retain[s] certain data from your interactions with us, but [they] take steps to reduce the amount of personal information in our training datasets before they are used to improve and train our models.” (How your data is used to improve model performance. (2025). OpenAI.com. https://openai.com/policies/how-your-data-is-used-to-improve-model-performance/). However, there is no explanation in the article as to what steps are taken or what amount of “personal information” is considered to be “reduced” sufficiently that it can be used in their models.

Several AI tools have default settings that allow user data to be shared by default. Users need to opt out to change this setting. Some tools don’t give users the choice to opt out of sharing their data at all. Many of the AI tools suggest going into their settings and modifying things such as chat and privacy functions in order to have some sort of sense of control over how much of your data is shared. However, making these changes once is probably not enough. New default settings are often rolled out, so your preferences may be changed and updated to new default settings without your knowledge. The main idea is that you should assume that anything you type into ChatGPT will be publicly available afterwards, because there are no protections against such a thing.

Another major concern with AI and the legal field is the accuracy of information being churned out. There have been issues in multiple jurisdictions of lawyers using AI to write briefs and running into disciplinary issues because the AI tool they used cited a completely fake case. An AI tool will make up a legal authority and present it as authentic. AI tools have also been shown to be typically inaccurate in complex legal analysis. Further, because the algorithms used by AI were created by humans, studies have shown that AI models tend to demonstrate bias against certain groups of people based on the results generated from the algorithms. For these reasons, AI should not, under any circumstances, be considered a reliable source of legal advice. If you need legal advice, you should always consult an attorney.

As a client, you should be aware of whether the use of AI is covered by your attorney’s fee agreement. Many law firms are incorporating the use of AI into their standard fee agreement to establish expectations and protect client confidentiality. There may be rules guiding the input of case information into AI by both the client and the law firm, depending on the agreement.

It may be tempting to ask AI a legal question, or to ask it to draft a document for you. However, you should be aware that there are many potential pitfalls to the situation, including privacy and accuracy concerns. The best source of legal advice is always an attorney. If you are looking to speak with an attorney, please contact our office to schedule a consultation.