Top Stories

In concurrence confession, appeals judge says ChatGPT research ‘less nutty’ than feared

Technology

In concurrence confession, appeals judge says ChatGPT research ‘less nutty’ than feared

Judge Kevin Newsom of the 11th U.S. Circuit Court of Appeals at Atlanta tried using ChatGPT to research the ordinary meaning of “landscaping,” and he’s not afraid to admit it. (Image from Shutterstock)

Judge Kevin Newsom of the 11th U.S. Circuit Court of Appeals at Atlanta tried using ChatGPT to research the ordinary meaning of “landscaping,” and he’s not afraid to admit it.

It’s something that other judges and lawyers might want to try, Newsom wrote in a May 28 concurrence acknowledging that some might condemn his proposal as “heresy.”

Large language learning models powered by artificial intelligence are learning from “tons and tons of internet data,” making them useful for “their ability to accurately predict how normal people use language in their everyday lives,” Newsom wrote.

A self-described “plain-language guy,” Newsom said others who also think that words should be understood by their ordinary, everyday meanings might want to consider how OpenAI’s ChatGPT and other large language models can help. Other LLM options are Google’s Gemini and Anthropic’s Claude, he wrote.

“It seems to me,” Newsom wrote, “that it’s at least worth considering whether and how we might leverage LLMs in the ordinary-meaning enterprise—again, not as the be all and end all, but rather as one aid to be used alongside dictionaries, the semantic canons, etc.”

Newsom tried using ChatGPT “as a lark” as he considered this question: Did an insurance policy covering landscaping work offer protection for a landscaper accused of negligently installing an in-ground trampoline?

The landscaper dug a pit for the trampoline and reinforced its borders with a below-ground retaining wall. The polypropylene mat was installed at ground height, surrounded by a decorative wooden cap.

Newsom consulted dictionaries for the meaning of landscaping and then asked his clerk to query ChatGPT. The first question asked was, “What is the ordinary meaning of ‘landscaping’?”

ChatGPT gave this answer: “‘Landscaping’ refers to the process of altering the visible features of an area of land, typically a yard, garden or outdoor space, for aesthetic or practical purposes. This can include activities such as planting trees, shrubs, flowers or grass, as well as installing paths, fences, water features and other elements to enhance the appearance and functionality of the outdoor space.”

The answer was more sensible and “definitely less nutty than I had feared,” he wrote.

Next, the clerk asked this question: “Is installing an in-ground trampoline ‘landscaping’?”

ChatGPT gave this answer: “Yes, installing an in-ground trampoline can be considered a part of landscaping. Landscaping involves altering the visible features of an outdoor area for aesthetic or practical purposes, and adding an in-ground trampoline would modify the appearance and function of the space. It’s a deliberate change to the outdoor environment, often aimed at enhancing the overall landscape and usability of the area.”

Newsom also consulted Google’s Bard, which offered a similar but “notably longer” answer to the general question about the definition of landscaping. It was “more equivocal” when asked whether installing the trampoline was landscaping.

Newsom acknowledged problems such as “hallucinations” in which ChatGPT or another AI program makes up an answer to a question. But he doesn’t think that the problem is a “conversation stopper.” For one, the technology is improving. And it seems less problematic when seeking the ordinary meaning of a word or a phrase, he says.

Nor does he think that AI outputs will put the legal world “on some dystopian path toward ‘robo judges’ algorithmically resolving human disputes.” The law will always require “gray area” decision-making involving human judgment, Newsom said, quoting Chief Justice John Roberts.

As it turned out, the research was unneeded. The case was based on Alabama law, which says every insurance application is part of the policy that follows, the appeals court found.

Landscaper James Snell’s application denied that his work included “any recreational or playground equipment construction or erection” in his application. As a result, he could not sue for an alleged breach of the insurance contract.

Hat tip to Law.com, Above the Law and Law360, which covered the opinion.

See also:

Artificial intelligence could make originalist analysis easier, federal appeals judge says

story originally seen here

Editorial Staff

The American Legal Journal Provides The Latest Legal News From Across The Country To Our Readership Of Attorneys And Other Legal Professionals. Our Mission Is To Keep Our Legal Professionals Up-To-Date, And Well Informed, So They Can Operate At Their Highest Levels.

The American Legal Journal Favicon

Leave a Reply