If you’ve been following the news recently, you’ll know that Artificial Intelligence (AI) is a hot topic these days. AI means a demonstration of intelligence by a machine that is programmed to understand, infer and synthesize information. We are seeing news of AI regularly now (the AI chatbot ChatGPT being the most recent example). It’s clear that AI has many potential applications for humanity, including advanced web searching, self-driving vehicles, and medical diagnoses.
As more and more jobs are automated with the help of AI, a natural question is, are there some jobs that just can’t be done by a machine? What about a lawyer’s job, can that be performed by AI?
One company is trying to answer this question using AI software developed to handle a speeding ticket for two individuals. The software company, DoNotPay, has promised to cover the costs of the ticket if the judge rules against the persons who received the tickets. DoNotPay has developed a chatbot that will suggest legal responses in court, which the individuals can then choose to repeat to the judge. The CEO of the company argues that the high cost of legal fees restrict access to legal counsel for many people, and an AI lawyer would help alleviate that problem.
While a fascinating concept, critics argue that bringing an AI into the courtroom would constitute the unlawful practice of law, as only licensed attorneys and a limited class of others can offer legal advice to people. In addition, privacy advocates have concerns about the use of AI, which must record audio from the courtroom in order to work correctly (most courts have strict rules against audio/video recording).
From a lawyer’s perspective, there are also real ethical concerns over confidential client communications. If a client tells an AI chatbot something sensitive and confidential, who (or what) has access to that communication? How will a computer negotiate the ethical rules and guidelines placed on human lawyers to ensure that client confidences are held sacred?
Lawyers are one of the few professions that self-regulate; that is, lawyers set rules for themselves. I suspect that if AI were to gain wider acceptance in courts, it would only be because human lawyers have agreed to allow it. Before that could ever happen, AI would need to demonstrate that it was at least as good, and likely even better, than a human practitioner. I don’t see that seriously happening anytime soon, and so for now, AI lawyers will remain a novelty with the potential to alter the legal landscape in the future.