Recently, the use of Artificial Intelligence or “AI” has become a topic of conversation at the workplace, dinner table, and even the beauty salon. While it has always been understood (and begrudgingly accepted) that technological advances result in less need for skilled laborers, the use of AI has brought on a whole new fear: the loss of the need for human assistance in general.

While this seems like an impending disaster, I am here to tell you that it will be a while before AI takes over the world.

In its current stage of development, many companies who specialize in machine learning and/or data analysis are hiring teams of experts across various professions to train AI chatbots. Training includes feeding the bot a prompt, which could be a question or a command, and thereafter analyzing its responses. If an acceptable response is not generated, the professional has the task of assisting the bot with developing the perfect response. However, given technical glitches and human fallibility in general, this hasn’t proved to be an easy process.

Think of an AI system as a large database. The database can be used to store information, but can also be programmed to retrieve information as well. It would take an immense amount of time and configuration to upload, review, and develop the proper algorithms to train the system to produce the perfect response. In doing so, human assistance is entirely necessary. The system cannot learn to be like US, without US. There are many aspects of humanity, such as the ability to process logic and reason, in conjunction with the ability to feel and express emotion, that cannot be adequately programmed into a machine without detrimental effects to society and societal systems.

What Does This Have to Do With Law?

You may have heard in recent news that a lawyer used an AI-based service called “ChatGPT” to draft a document that was later filed with the Court. Unfortunately, it was discovered soon after that the cases cited in the filing were not on point. In fact, they didn’t exist at all.

In what has been referred to as an “AI hallucination,” the bot will include case law, complete with page citations, to a document in an effort to satisfy the user request regardless of whether said case actually exists. The issue was discovered when opposing counsel was unable to find and review the cited law to verify the holding. Naturally, this has brought on concerns about the use of AI in the legal profession, its shortcomings, and whether lawyers should be sanctioned when they inadvertently submit false information to the Court.

That being said, the use of AI in law is likely a concern to be addressed by the ABA and other state ethics boards in the near future. Under the ABA Model Rules of Professional Conduct:

“a lawyer shall not knowingly make a false statement of fact or law to a tribunal or fail to correct a previously made false statement of material fact or law… previously made by the lawyer.” Model Rules of Professional Conduct R. 3.3 (2023)

Traditionally, all filings submitted by the lawyer bearing their signature are viewed as statements by the lawyer. When you file a document, you are knowingly asserting to the court that the information contained therein is true to the best of your knowledge. In the case of the unchecked use of AI, it is difficult to say whether the actual misrepresentation of law was made knowingly. However, lawyers are also ethically required to do their due diligence in obtaining accurate information. As it stands, the use of AI in the legal profession invokes several ethical rules which may necessitate a new rule altogether. In light of this, it is imperative that members of the legal profession exercise caution in using AI based programming to increase efficiency. Due to the shortcomings of AI programs and the continued need for training, analysis, and re-training, it doesn’t appear likely that we will be replaced by robots any time soon.