David Zvi Kalman is the host of the podcast “Belief in the Future,” a fellow at the Shalom Hartman Institute, and a senior advisor at Sinai and Synapses. Kalman is the owner of Print-O-Craft Press, an independent publishing company, and his own work is widely published in scholarly and popular publications. He holds a PhD from the University of Pennsylvania and a BA from the University of Toronto. He blogs regularly about Jello Menorah.
Religious thinkers around the world are currently trying to figure out what to do about AI, albeit in a haphazard way. A meeting has been called. We also accept requests for books and articles. Google “Muslim approaches to AI” or “Catholic approaches to AI” and you’ll see what they actually say. Most of these have happened in the last five years. The wheels are really turning.
But as someone who has been watching this closely, I can tell you that everything is very slow, which is a huge problem in an industry that values speed above all else. In a world that is moving faster than ever, religion cannot afford to slow down.
What I find slow is that I still can’t answer the following questions: Muslims/Catholics/Jews/etc. Are the approaches to AI essentially the same, or are they significantly different? If so, what exactly is the difference?
This is an important question to explore because it forces religious thinkers to get serious and decide whether they want to be ethical leaders or ethical followers. It’s easy to say that your beliefs stipulate, for example, that AI should not be used to exploit the weak and vulnerable. In fact, it would be strange if it wasn’t. Because literally everyone agrees with this, religious or not. Religious sects that promote such consensual positions are unlikely to make any enemies, but they do run the risk of getting lost in a sea of nodding opinions. In fact, if everyone wants the same thing, using religious language to say it is often perceived as an obligation. Because such words can be alienating in a secular society.
This is why I am drawn to disagreements. This means that some religious groups are saying things about AI that others are not. As a Jewish thinker who has worked on technology issues and spent time talking with AI ethicists from other religions, here is where I believe Jewish emerging AI ethics is doing something distinctive. I’ll list it.
“Godlikeness” is not just about biology
AI has demonstrated human-like capabilities for any task. Concerns about the erosion of human values have arisen because humans understand their worth in relation to their unique abilities. Both Christians and Jews have responded to this alarm by emphasizing the religious idea that humans have a special status as unique creatures created “in the image of God” (Genesis 1:27). ).
However, not everyone uses this phrase the same way. The Southern Baptist Convention, for example, understands this to mean valuing biological human beings, including unborn children, above all other beings. This understanding motivates the sect’s opposition to IVF and abortion, as well as its categorical denial of personhood to all AI, no matter how human-like it becomes. Humans are simply ontologically separate from both animals and machines.
As I understand it, Jewish thought follows a different path. To be sure, biological humans have infinite value, but that value is also tied to what we perceive to be human by the degree of personhood available in marginal cases. This approach lies behind Judaism’s complex position on abortion, among other things. In my opinion, this also means that we should seriously consider assigning some degree of personality to machines that behave like humans. While some may have the knee-jerk reaction of thinking that this violates human rights, I believe the opposite is true. If we think that human capabilities are “only” equivalent to what AI can achieve, we are more likely to disrespect humans. The ability of AI to behave like humans should enhance it, not detract from it. Similarly, intentionally creating human-like software is different from creating other types of software. Just as AI companies explicitly compare their products to nuclear technology in terms of its promise and danger, we recognize that it shares something with the creation of human life. developers must understand this as an obligation to develop and maintain the model. be careful.
The obvious drawback of this approach is that it can be used to dehumanize people with disabilities. For this reason, it is important to emphasize that we understand the Jewish approach to be a hybrid. Both biological human nature and human behavioral patterns contribute to the way we value living things and machines. This perspective, so to speak, has not been crystallized, but the differences are already becoming clear.
The role of AI in religious text research
In Judaism, the study of texts is not just a means to understanding tradition. It is a ritual act in itself, and all Jews (although there are actually far more men than women) are encouraged to pursue it throughout their lives. This desire to study difficult texts written in Aramaic, pre-modern Hebrew, and other languages led to a uniquely productive relationship between Torah and computing. Some of Israel’s first computers were conscripted to digitize the canon and make it easily searchable, and the number of digital tools now available to consumers is remarkable. Today, a vast number of religious texts are freely available online, with many translations and easy hyperlinks to navigate this deeply intertextual corpus.
The application of AI to Jewish learning is likely imminent. Generative AI already provides more than adequate translations of even the most difficult texts, helping learners navigate large and unwieldy corpora with few clear entry points. In the future, AI may allow learners to communicate directly with rabbis and centuries-old books.
At the same time, I suspect that the ritual aspects of Jewish learning mean that AI will enter the Beit Midrash (hall of learning) without complete conquest. The ability to read the Bible and Talmud in their original languages is likely to remain a cornerstone of text study, as is hebrta, the act of pouring over a text with a partner rather than alone in front of a screen. Jewish learning also emphasizes the interaction between scholars and their interpreters. AI models, on the other hand, often struggle to explain why they know what they know. Most importantly, studying the Torah emphasizes that sacred texts are often multivalent, layered, and mysterious. This way of presenting text is difficult to break away from an AI model that is primarily designed to provide a concise and ostensibly complete interpretation of the source material.
In practice, I think this means that AI will become another technology in a series of technologies that have enriched Jewish learning while leaving its predecessors intact. . Modern midrash includes handwritten scrolls, bound books, and computer screens. Each has a place, and learners move between them as needed. It is easy to imagine that AI will be added to this.
AI and Jewish Theology
AI has long stimulated religious sentiments in people. In some cases, it is because a powerful yet poorly understood model appears to be god-like. In some cases, it happens through “singularity,” which is already a quasi-messianic concept. Sometimes it is because the act of creating something in our image is reminiscent of the sacred act of creation.
For monotheists, this last sentiment is powerful, but difficult to think through to its conclusion. If creating AI is like creating humans, if creating new forms of life replicates divine acts, then it seems to make us gods. In fact, Yuval Noah Harari and others use these very words to describe the trajectory of technological development. Is it possible to understand AI as a descendant of humanity and avoid this language?
I think the answer is no. AI is nothing more than a radical rethinking of Jewish theology as a kind of precedent. In it, the long history of humanity’s relationship with God is reexamined in light of our current experience of creating something in our image. Both AI as God and AI as Son of Humanity contain notes of potential heresy, but the former is far more dangerous ethically and religiously. I think the choice here is obvious.
What’s important about this parallel is that both humans and AI are concerned that their descendants will go off track and that their “code” (in both senses of the term) is not clear enough. It is something that is causing extreme concern in our hearts. This concern, which developers call an “integrity issue” and Jews call “all of history,” should generate sympathy for God among religious people. Even if we believe that God is perfect, it is clear that there is no such thing as a perfect code. The Hebrew Bible itself contains countless examples of how humanity repeatedly struggles to do the right thing long after the revelation at Mount Sinai. We also hope to evoke real anxiety in those developing AI systems. AI is not just a new technology. It is a technology that has the potential to define who we are as a species.
Convergence or divergence?
Does the above define a unique “Jewish perspective” on AI?
Of course not. Judaism is a decentralized religion, and most Jewish leaders (like most religious leaders) do not understand technology well enough to take informed positions. The process of spreading AI literacy and building consensus will take some time. But while the world’s religions will largely converge in how they think about AI, just as they do about the need for climate protection, I think they will still remain different in important ways. . Those differences may seem small now, but like the Jewish decision to keep handwritten Torah scrolls on the surface of a printed book, or the choice not to use electricity on the Sabbath, In the end, they may become the defining characteristics of religion. Religious thinkers of all faiths should keep this in mind when formulating their approaches.
