“Artificial Intelligence is rapidly becoming a ubiquitous presence in our daily lives, with its increasing popularity transforming the way we live, work, and communicate. From virtual assistants to self-driving cars, AI is revolutionizing industries and offering new possibilities to tackle complex problems.”
Who said that? Here’s a hint: It wasn’t a local tech expert or classroom teacher. That paragraph right there was written by a popular A.I. web tool called ChatGPT in response to the prompt “write a newspaper lede about the growing popularity of AI.” (The system even knew that “lede” is an industry-specific spelling of the word “lead,” meaning the first sentence in a news story.)
Here on the South Fork, people can’t help but be curious about its potential; some are even experimenting with using it professionally. Among those people is Rabbi Joshua Franklin of the Jewish Center of the Hamptons, who took to ChatGPT to “write” a sermon in December, then had his congregation guess its source. This created such a ripple that the story was picked up by both religious and secular news outlets, including Semafor, The Times of Israel, Fox News, BBC, The Jerusalem Post, and The Forward.
ChatGPT “was still really in its infancy, and I had been reading a couple stories here and there about the massive abilities of ChatGPT to take on novels, speeches, and homework assignments,” Rabbi Franklin said in an interview. “Would it be able to do some of the tasks that I normally do as a rabbi, particularly sermon writing? The answer is yes, it can do those things. It wasn’t the greatest sermon — I’m sure that will get better — but it had people fooled.”
He said he wanted to “give the congregation something novel. Give them not just what I normally do, but have a discussion about the implications of what artificial intelligence and ChatGPT mean for our future. There are a lot of scary but moving thoughts out there.”
For those who are imagining a dark future in which humanity lives in the shadow of A.I., like the way the 1991 film “Terminator 2” portrayed it, don’t worry. We’re not there. We are, however, at a critical point in time where “the issues surrounding A.I. are massive,” according to David Beard of Sag Harbor, who has a background in finance and research and is the president and chief executive officer of Bill Miller and Associates Inc., a tree care company in Water Mill.
“The impacts are likely to be the biggest in ways we currently cannot easily see,” Mr. Beard said in an email to The Star. “It’s easy to see that computer code or legal arguments can be run through A.I. and come up with an answer better than ‘ad men’ or ‘paralegals,’ but I am not sure we are going to ‘get rid’ of either any time soon.”
In writing his Jan. 2 “Corona Tracker,” a weekly report analyzing the pandemic and its impact, which he disseminates to friends, colleagues, and others, Mr. Beard employed a strategy similar to Rabbi Franklin’s.
“During my testing, I asked ChatGPT to summarize things that I felt I knew reasonably well,” Mr. Beard said. “I noticed that ChatGPT was very balanced in most of [its] responses. I also noticed that ChatGPT can draw upon a wide range of sources when developing its response. . . . I was not surprised by the results; they all ‘made sense’ and were balanced in their response — definitely not the response that gets clickbait or zillions of likes, if it were posted on social media.”
Mr. Beard asked the bot, “What should a business do to prepare for a recession?” It generated suggestions such as “review and assess your financial situation,” “identify cost-cutting measures,” “build up your cash reserves,” and “diversify your revenue streams.” Mr. Beard then identified the source of that advice.
“Yeah, they are rather bland and formulaic, but they were completely unedited and were the ‘first run,’ without modifications to search terms,” he concluded. “Pretty powerful indeed.”
In schools here, administrators are keeping their eyes on A.I. and news reports elsewhere of students using it to complete their assignments. Neither East Hampton High School nor the Bridgehampton School has dealt with it firsthand yet, said their principals, Sara Smith and Michael Miller, respectively.
“That is not to say that we are immune to the problem,” Ms. Smith said in an email, “but [our teachers] have not had a specific example of a student using A.I. technology and claiming that work as their own. To ensure academic integrity and honesty, larger and more complex writing assignments are often done in class where students have access to any necessary supports, thus not having to resort to other, less ethical, means. Additionally, we are at a point in the year where our teachers have learned each student’s writing style and sense of voice. That means they are able to tell when a student’s writing is authentic and when it is not.”
Mr. Miller said in a Jan. 23 email that in Bridgehampton, “we are aware that technology continues to improve outcomes for all stakeholders, while at the same time it can be used for disingenuous reasons. Our technology support team continues to monitor in order to ensure the safety of all staff and students. With that said, A.I. technology, as of today, has not been a concern in our district or from parents/community members. . . . If technology can be utilized to assist individuals in a positive, productive, and legal manner, than by all means we should find ways to assist people where necessary.”
What did Rabbi Franklin learn from the experience? Religious leaders including rabbis are not going to be “the continuing purveyors” of written and spoken material. “Instead, we’re going to really learn to hone other qualities that make us who we are,” he said. “Compassion, love. We’re going to have to work on developing relationships and community more strongly. That is at the core of what we do. ChatGPT can’t synthesize those things.”
When a user logs in to the ChatGPT website, which right now allows research inquiries for free, a warning immediately pops up: “While we have safeguards in place, the system may occasionally generate incorrect or misleading information and produce offensive or biased content. It is not intended to give advice.”
Now, about that newspaper lede.
“With its ability to analyze massive amounts of data, make predictions and automate tasks, AI is expected to significantly change the future of work and create new job opportunities,” ChatGPT continued. “As the technology continues to advance, its potential for both positive and negative impact on society is increasingly coming under scrutiny, raising important questions about ethics, privacy, and the role of AI in our lives.”
It’s not wrong, but it’s not going to win a Pulitzer Prize, either.
