Menu
Menu
December 17, 2022
Author(s): Rabbi Michelle Robinson,
Parshat Vayeshev
ChatGPT and Me
December 17, 2022 — 23 Kislev 5783
Temple Emanuel, Newton, MA
On Wednesday night, I came home after 10 PM only to hear raucous laughter coming from Eli and Benjamin’s room. As I approached, the door opened, and there was Mike laughing too. In very helpful spouse mode, I asked, “What in the world are you doing that’s keeping them up so late on a school night?” In unison, they replied, “ChatGPT!” Chat what? I asked.
If you’re in the know, you may be fired up by ChatGPT, which has hit the tech world like wildfire. If you’re like me, you may be wondering, what exactly is this thing that’s getting so much buzz?
Wikipedia describes ChatGPT as “a prototype artificial intelligence chatbot developed by OpenAI which specializes in dialogue.” That may not sound like much. But that “dialogue” has the potential to literally transform our world. Why? Well, it has to do with the fact that if you feed the program the right assignment, it produces a reply that can be indistinguishable from one penned by a human hand.
Testing the boundaries of that can be so engrossing that soon Mike was not alone in keeping the boys up late. They shared with me the amazing verse spit out by their random online prompts. “Write a sonnet about blue hair” resulted in a stirring, “With tresses dyed a shade of azure blue, you stand out in a crowd a shining star, your hair a canvas for your artistic view, a reflection of the person you are.” “Write a paean to the number zero” approached the mystical, “In your infinite nothingness, you contain the potential for everything. You are the space between, the silence between sounds, the pause between thoughts.” On and on, every writing assignment resulted in a reply cogent and creative enough that I almost felt like shaking the phone to check if there was a little person inside.
What’s more – this is just an early model. Imagining that the technology improves, as it certainly will, we may be within reach of an artificial intelligence that can convincingly and consistently pass what is called the “Turing test,” communicating in a manner equivalent to and indistinguishable from a human. That is why, according to those in the know, this is ground-breaking technology. Think back to a world before smartphones, before the internet, and how much the world has changed since. Have any of you used a library card catalogue recently? No? They were rendered obsolete by the powerful search engines we all now have in our pockets. Which means that, where ChatGPT is concerned, just like the advent of any world-changing technology, there are avid enthusiasts and concerned critics.
Critics call out the destabilizing allure of authoritative-sounding disinformation, of intellectual arguments unconstrained by ethics. Ezra Klein points out in the New York times, “A.I. systems like this make the production and manipulation of text (and codes and images and eventually audio and video) functionally costless. They will be deployed to produce whatever makes us most likely to click. But these systems do not and cannot know what they are producing.”
Educators fear what happens when kids can input a query to a system which spits out a full-blown analytic essay. Why would any teen do their own History or English homework when simply typing in the prompt, “Compare the impact of the 1918 flu pandemic to Covid-19” or “Write a literary analysis examining the theme of isolation in Mary Shelley’s Frankenstein” results in a fully formed response?
Doctors wonder and worry about the abilities of such a system to diagnose – or mis-diagnose – medical ailments without a human hand. Knowledge workers fear they may be out of a job. Most distressingly, there’s the all-too-human propensity of ChatGPT and its A.I. siblings to flat-out lie. As Charles Seife shares in Slate magazine, when he challenged an OpenAI chatbot named Davinci-003 to write his own eulogy, it fabricated not just facts but references – a human attribute that I am confident most of us would agree the world does not need more of.
Enthusiasts, on the other hand, point out boundless possibilities. Zeynep Tufekci wrote Thursday in the New York Times, “Plato mourned the invention of the alphabet, worried that the use of text would threaten traditional memory-based arts of rhetoric… that it would impart ‘not truth, but only a semblance of truth.’” She goes on to imagine a world in which students would use ChatGPT in much the same way that math classes use calculators, as an aid to higher-level learning. “[T]he ability to discern truth from the glut of plausible-sounding but profoundly incorrect answers,” she argues, “will be precious… The way forward is not to just lament supplanted skills…but also to recognize that as more complex skills become essential, our society must equitably educate people to develop them.”
In other words, the technology itself, like new technologies since the dawn of time, has the capacity for great good or great evil; it all depends on what we do with it. Which is an ancient truth that emerges from this week’s Torah of Joseph and his dreams. When we first encounter Joseph, he is a narcissistic know-it-all who shares his dreams with a seemingly complete absence of EQ. That his brothers are jealous, that they hate him, is clearly and repeatedly articulated by the text. But none of that phases Joseph. Like early A.I., he seems to produce the words with reckless lack of regard for their impact.
Of course, we know how that goes down. His brothers’ anger boils over, and a tale of woe unravels.
But what ultimately happens to Joseph in Egypt? The rest of the story is not about Joseph’s dreams, not about his own content-production, but about his capacity to interpret the dreams of others. It is there, in looking beyond himself, bringing his analytic capacity to interpret what is before him, that Joseph discovers his true purpose. And so, perhaps, can we.
I have no idea if one day we will look back at the dawn of ChatGPT as the gateway to dystopia, utopia, or something in between. But I do know that the rabbis of the Talmud taught that “dreams go according to the interpretation.” Which means that we are well-advised to meet our anxiety with inquiry. The seeds of our future story lie in the interpretation and applications we bring today.
With that in mind, I thought it would be intriguing to ask ChatGPT to write a sermon comparing itself to Joseph and his dreams. Here is what ChatGPT wrote. I will leave it to you to interpret:
“Dear brothers and sisters,
Today, I want to talk to you about two important figures: ChatGPT and Joseph. You may be familiar with ChatGPT, as it is a powerful and advanced language model that is capable of generating human-like responses to a wide range of questions and topics. But you may be less familiar with Joseph, who is a figure from the Bible known for his ability to interpret dreams.
In the story of Joseph, we see how God used his gift of dream interpretation to guide and protect him, even in the midst of difficult circumstances…
But Joseph’s dreams were not just about his own future; they also had a deeper significance. In his interpretation of Pharaoh’s dreams, Joseph revealed the Lord’s plan for the land of Egypt and saved the people from famine. This shows us that God can use our gifts and talents, no matter how small or insignificant they may seem, to accomplish great things and fulfill his purposes.
In a similar way, ChatGPT is a powerful tool that can be used to generate human-like responses to a wide range of questions and topics.
Like Joseph, it has the ability to interpret and understand complex ideas and provide valuable insights. But, just as with Joseph, it is important to remember that ChatGPT is not a replacement for human intelligence or wisdom. It is a tool that can assist us in our pursuits, but ultimately, it is up to us to use it wisely and in accordance with God’s will.
So, let us remember the lessons of Joseph and his dreams as we continue to use and develop ChatGPT. Let us use it to further God’s kingdom and to help others, always staying true to his word and guidance. May God bless us all. Amen.”
Thank you, ChatGPT. I could not have said it better myself.
Shabbat Shalom.