Artificial Intelligence and the Practice of Law Part 2: Working With Your New AI Staff Attorney

This article discusses how artificial intelligence (AI) can assist lawyers in their legal practice. The article explains how generative AI systems work, using large language models and transformer technology to learn from the data on the internet and produce responses to user prompts. The article also provides some advice on how to best use AI in legal settings, such as when to use AI for background information, when not to use AI for finding specific sources, when to use AI to confirm what you already know, and when to use AI to digest, analyze, or summarize authorities that you have verified. The article cautions against relying on AI for complex legal research and analysis, and warns of the risks of AI hallucinations, semantic confusion, and jurisdictional errors. In particular, the article discusses:

• Generative AI systems can generate expressive content such as text and images based on user prompts and data from the internet. Generative AI systems use large language models and transformer technology to learn from the data and produce responses.

• Generative AI systems can be used for background information, factual research, and summarizing or explaining information in simple terms. Generative AI systems can be used to confirm what you already know, but not to perform complex legal research and analysis in areas where you are unsure of the answers and sources.

• Generative AI systems have limitations and risks, such as hallucinations, semantic confusion, and jurisdictional errors, that require constant supervision and verification and citation checking by human lawyers.

The article concludes by suggesting that AI can be a valuable staff attorney if used in appropriate situations and under constant supervision by human lawyers.