Writing drashot is hard, particularly for the High Holy Days. I’ve been doing this for a while now, 17 years I think more or less. Times four, that’s a fair number of High Holy Day drashot over the years. Plus, remember I write shorter ones, twice a month during the year. At some point I get tired of hearing myself think out loud. So I thought, why not let ChatGPT do the heavy lifting this year? After all, students are doing it now and academics have probably been using it for years, since most of their writing is unintelligible anyway. But it’s Rosh Hashanah and I shouldn’t be mean so I’m going to have that last laugh line stricken from the record.
So I created a ChatGPT account, the free one of course, and wrote a prompt, “Write a sermon about Abraham binding his son Isaac. How it was sublime proof of his deep faith.” And I got this for the opening paragraph, “Dear brothers and sisters, today we gather to reflect upon one of the most profound and challenging stories of faith found in the pages of our sacred scriptures – the story of Abraham binding his son Isaac. This tale from the Book of Genesis not only reveals Abraham’s unwavering devotion to God but also offers us valuable lessons about faith, sacrifice, and the divine plan that shapes our lives.”
I thought, this is a little to Christian sounding so I wrote another prompt, “Write the same sermon but within a Jewish context,” This time I got, “Shalom, dear members of our community. Today, we gather to delve into a poignant narrative from our sacred Torah, a story that resonates with the essence of Jewish faith – the Akeidah, the binding of Isaac. This story found in the Book of Genesis not only highlights Abraham’s unwavering devotion to Hashem but also offers profound insights into the power of emunah, faith, that forms the bedrock of our tradition.”
That sounded a little better. I appreciated the sprinkling of Hebrew… you know, shalom, Torah, HaShem and so on. But in my opinion it should have written Sefer Bereshit, not the Book of Genesis, but perhaps Chat GPT is not as Jewishly literate as I had hoped. Also, something else was missing. You all know that I frequently like to begin my sermons with a joke or a personal anecdote to draw you all in, so I wrote, “Keep the Jewish sermon but open with a joke.” I got, “Shalom, wonderful members of our community! Before we delve into our sacred teachings, allow me to share a lighthearted Jewish joke that beautifully sets the stage for our discussion today.
Joke: Why did Isaac bring a calculator to Mount Moriah? Answer: Because he knew it was time for some serious “counting” on God’s promises!
Now, with a smile on our faces, let us turn to a profound narrative from our Torah, the Akeidah, or the binding of Isaac. This story not only tickles our sense of humor but also profoundly touches on the core of our faith – the enduring power of emunah, our unwavering belief in Hashem.” So first of all, what makes this lame joke Jewish? Sounds more like an accounting joke. Plus, this story “tickles our sense of humor?” You can say many things about the Akeida and believe me I have over many Rosh Hashanah mornings. But, tickling our sense of humor is not a phrase I would ever apply to this story. Horrifying, yes. Difficult, yes. Incomprehensible, ok. But humorous, I don’t think so. That transition sentence was seriously lame… exquisitely bad. So, at least for the moment I am not worried about ChatGPT taking my job.
Then I decided to go for broke, because I would really like it if I could get a computer to write my drashot. So I wrote, “Write a Rosh Hashanah sermon, approximately 1,000 words, about what makes us human and whether AI can have faith and feelings. Tie it into Abraham binding Isaac. Make it sound like a rabbi wrote it.”
Here is the newest opening, “Shalom and blessings to all of you, my beloved congregants, as we gather on this sacred day of Rosh Hashanah to reflect upon the mysteries of existence and the profound questions that define our humanity. As we blow the shofar, its reverberations call us to awaken not only to the new year but also to the depths of our souls, where we seek to uncover the essence of what it truly means to be human.
In our Torah portion for this Rosh Hashanah, we encounter the story of Abraham’s binding of Isaac. A tale that has echoed through generations, this story prompts us to explore the nature of faith, sacrifice, and the unique qualities that distinguish us as human beings.”
Pretty journeyman like, but much better don’t you think? Here, in my opinion, is the essential paragraph, “This foundational tale teaches us that our humanity is rooted in our capacity for moral discernment, empathy, and choice. These are qualities that set us apart from all other forms of life. As we stand on the precipice of a new era marked by advancements in artificial intelligence, we are confronted with a profound question: Can machines possess the qualities that make us uniquely human?”
So at least the computer got the fundamental question: what makes us human and can we create a machine that emulates us? Of course, that wasn’t hard. It was what I asked it to do. But I think the computer, the AI, or the large language model, call it what you will, provided the wrong answer: both in the Jewish sense and in the truest sense.
First, in the traditional, Jewish view what makes us human is that we are created by God in the Divine image. Each of us is unique, creative, free to make choices, and endowed with inherent dignity as creations of God. Machines, of course, are creations of human beings. They are not even created in the human image, let alone the Divine one. They are mass produced, not unique, and have no inherent dignity beyond their utility, just like any tool such as a hammer or a saw. They are neither creative nor make choices. At least today. Will that change in the future? Perhaps, but who knows.
Our tradition is well aware of non-human beings. In the Mishnah, Masechet Kilayim (dealing with mixtures) Chapter 8, Mishna 5, it says, וְאַדְנֵי הַשָּׂדֶה, חַיָּה. רַבִּי יוֹסֵי אוֹמֵר, מְטַמְּאוֹת בָּאֹהֶל כָּאָדָם. Wild man-like creatures are [in the category of] hayyah. Rabbi Yose says: they cause impurity in a tent like a human being.” The Adnei Ha Sadeh, is some kind of wild human-like creature. The Tifferet Yisrael, a commentary on the Mishnah, says Adnei Ha Sadeh is an orangutan. Maimonides says that Adnei Ha Sadeh looks like a man, talks constantly, but is unintelligible. In the Ohr Zarua, he is described as appearing like a man in all ways, face, hands, and feet, but being of large size with an umbilical cord attached to the ground from which he draws his nourishment. If you want to kill it, sever the umbilical cord with an arrow (you don’t want to get too close). Further, it is written that there were soothsayers who could insert a bone in the mouth of Adnei Ha Sadeh and they could talk, while Rashi says that the Adnei Ha Sadeh could prophesy if the bone was put in their mouth. All very weird stuff, for sure. But, the important point is that our tradition believes that there can be something like a human being, who is not fully human. A missing link between humans and animals. Resembling a human being, but wild, and seemingly without volition.
Of course, most of us know about the Golem. The most popular story is of the Golem of Rabbi Yehuda Lieb of Prague, who creates a creature from clay, to protect the Jews from a pogrom and brings it to life by writing the word אמת (truth) on its forehead. But the Golem ultimately cannot be controlled and begins to run amok, so Rabbi LIeb erases the א from the Golems forehead creating the word מת (dead) and the Golem turns to dust. Jewish tradition suggests that we are all Golems in a way, for Adam, the first human is formed from clay, from soil (אדמה) and only comes to life when God breathes life, an animating soul, into him. Then Adam and Eve eat from the tree of knowledge of good and evil and they develop self-awareness, of themselves as separate creatures, of their bodies, of their desires, of their mortality. Until that moment they are more like animals than humans.
This seems to me to be the fundamental difference between human beings and all other forms of life, as well as machines: we alone are self-aware and the most important awareness may be of our mortality. All of us know that our time is limited and, sooner or later, we will die. As far as we know animals do not have this perception, they live in the moment (like Adam and Eve before they ate the fruit of the Tree of the Knowledge of Good and Evil) and theoretically speaking, machines don’t die. It is our self-awareness of the fleeting nature of our lives, the limited time we have on earth, that makes us human.
There is something else in the Golem story, and there are multiple Golem’s from Talmudic times through today, In every case where a human being tries to emulate God by creating artificial life, and that is what the Golem is, the creature ultimately goes beserk, beyond its creators control, and must be destroyed. This is a cautionary tale for us, about our technological reach going beyond our grasp. One that we consistently ignore. Since the Enlightenment, at least in the West, we have believed that technology improves the human condition, despite the Holocaust, where all the tools of Western technology, industrial production, engineering, efficiency, were employed for mass murder. Or the undeniable evidence of our destructive impact on the Earth’s climate balance. What of AI? The Golem story reminds us to ask: what are we creating, what are the implications, and should we continue? Just because we can, doesn’t mean we must.
The Golem also hints at a confusion we have about AI. The Golem emulates a human being, but it is not human. It has no moral core, no understanding of right and wrong. It has not eaten the fruit of the tree. Many of us are intrigued by AI’s intelligence, or its potential intelligence. How a computer can beat a chess master. How a computer can process information so much faster than we can. But in this infatuation with intelligence (or maybe just processing speed) we confuse two different concepts: intelligence and goodness. There are lots of very smart people who do very bad things. Among Hitler’s earliest and most enthusiastic supporters were academics, university students, and professionals. The SS was filled with Phds, Well educated people, who we think should have known better. Clearly education and intelligence is no guarantee against evil. The smart thing to do is not necessarily the right thing to do. I don’t know if it is smart to help the poor, or care for the weak and the disabled, or to show no special favor to the rich and powerful, but according to Jewish law it is the right thing to do. That is how we, and our society, are ultimately judged, not by how smart we are. Kindness and empathy may not be efficient, but they are what makes us good. Can a machine be good? Who cares how smart it is, if it has no moral grounding? Especially, if those controlling it have no moral grounding.
Our conversations about AI seem to be centered around how it will impact the labor market. Will it make lawyers and doctors more efficient, or unnecessary. Will students use it to write papers? What we’re not talking about, what we never talk about with emergent technology, is the moral implications, how it will impact us, how we treat each other. We have already seen the negative effects of social media on our political life, on our children’s mental health. Did we not see this coming or did we choose not to think about it?
At this point you may be wondering why I’ve been talking about AI and not the Torah portion, or maybe not, but I do feel some obligation to do so, not the least because this is the assignment I gave ChatGPT, and because we’re in the Torah service and the traditional purpose of a drash is to provide some insight into the Torah portion.
So here is what ChatGPT got wrong about the Akeida. What makes it so powerful, no matter how often we hear it, is that we understand, viscerally, that God is demanding Abraham perform an impossible task. A demand that God should not be making. No God should demand a parent murder their child. Who is capable of such an act? This is the demand of a tyrant, not a God that is the manifestation of ultimate morality. And Abraham agrees: immediately, without hesitation. This is why the story is so unsettling. For millennia, apologists, Jewish and non-Jewish, have argued that this is a demonstration of Abraham’s faith, his complete trust in God; just as ChatGPT put it. Our sages also argued that this story proves that God does not want children to be sacrificed to him, as was supposedly common practice in the ancient near east and that it took such a soul shattering story to bring this change about.
I don’t buy it. Rather, I think this is where our awareness of our mortality comes in. Sarah and Abraham have been waiting their entire lives to have a child. Isaac is the child of their old-age. He is the one through whom all their hopes and dreams will come true. Then God asks them to destroy their future and with it all their hopes and dreams. We all know, in the deepest recesses of our hearts and souls, this is an unreasonable demand, because we know about hopes and dreams and the fleeting nature of life. God may be able to see all eternity, but we cannot. We just see our little slice of it: our lives, our children’s lives, maybe our grandchildren’s lives. That is how we view the world. This is where our humanity is forged and where our morality is formed.This is also the source of our empathy: from this place we can imagine Abraham’s internal suffering, Isaac’s fear, and (unspoken but real) Sarah’s shock.
What is the outcome in the end? Isaac is spared. The angel speaks. The ram is seen caught in the thicket. And… God never speaks to Abraham again, although he lives another 75 years. Sarah dies at the beginning of the next chapter. Isaac’s perception is forever altered. He never sees clearly again, for the rest of his life. What an absolute disaster. By the way, ChatGPT mentions none of this. It is a poor scholar of Biblical interpretation, but I digress.
As I share these pieces of the story with you, some are present in the text, some are present by their absence from the text, I hope you can feel every emotion of this story. One moment, one act, leads to such utter devastation in our founding family and echoes across time until today. This is not a function of intelligence or knowledge, or processing speed, but a function of emotions, heart, and our ability to relate to the suffering of another.
Perhaps this is why we read this passage on Rosh Hashanah morning, to remind us that life puts us in difficult places. That we sometimes have to make difficult decisions and the place to make them is from the heart; from the place where we can empathize with another. Not in certainty…, but in doubt. Not in absolute faith… but in questioning, in listening to our heart and conscience. It’s the best we can do, from the limited perspective that we have. It is what makes us human.
And even Abraham made mistakes, profoundly wrong judgements, as do we. Each year Rosh Hashanah comes around again with the chance to acknowledge our mistakes and change… if we want to. It is not about intelligence, it is about our shared humanity, our empathy, and our understanding that our lives are short and we can try to make the best use of them.