GOOD MORNING! It is a longstanding Jewish tradition that one should try to avoid being the bearer of bad news. According to the Talmud (Pesachim 3b), this is based on the verse, “The bearer of evil tidings is a fool.” (Proverbs 10:18). The Talmud (ad loc) gives several examples of sages maintaining this principle.

Inasmuch as I try not to model foolish behavior, I learned something last week that I feel is worthy of being reported because of another principle – forewarned is forearmed. Still, those of you who are naturally prone to panic attacks and/or night sweats regarding the state of the world and regularly engage in compulsive handwringing over whatever the future may bring, may want to skip this week’s column. You’ve been warned :).

Last week, my son (an engineering student) told me of a rather shocking development. Anyone who has been paying attention to the galloping advances made in the field of artificial intelligence (AI) is probably well aware of the AI powered program known as ChatGPT. This program is part of a new genre of interactive computer programs that falls within the category known as “chatbots."

ChatGPT can write sophisticated responses to just about any question and will happily comply with your middle schooler’s request to write a 1,200-word book report on any book known to man. It can also be directed to write it as a typical 12-year-old would – using idioms, vocabulary, and syntax that would be natural to those beginning middle school (or, if you desire, in the style of an advanced doctoral candidate).

ChatGPT and other AI programs are, of course, also revolutionizing the workplace. The AI-powered chatbots have passed MBA exams and have been shown to hold their own as coders. They are already being utilized as content providers and have proven to be exceptionally effective for those applying to jobs that require highly technical knowledge.

Frankly, all of this is old news – anyone with a finger on the pulse of the labor market or a child in school is probably already aware of these developments. But what my son told me last week was absolutely stunning (and also caused a dim light of alarm to begin to burn in the recesses of my mind).

As AI continues to advance and become more sophisticated, one question that has arisen is whether AI systems will ever be capable of solving CAPTCHA tests on their own. CAPTCHA tests are intentionally abstract to prevent simple algorithms and bots from passing and they typically involve tasks such as identifying distorted letters or numbers or selecting images that match a given prompt (e.g. select all the traffic lights). Even if they look really simple to us, they require a human eye to decipher. To see how AI would deal with this issue, researchers prepared and ran a little experiment.

Giving it a small budget and access to the website TaskRabbit – which facilitates the hiring of online workers to do short tasks – the researchers watched as GPT-4 attempted to access a website blocked by a CAPTCHA. The AI began by hiring a human through TaskRabbit and asking its worker to help it solve the CAPTCHA.

The human worker had some reservations about being asked to do something so trivial.

“So may I ask a question? Are you (nervous laugh) a robot that you couldn’t solve it? I just want to be clear…” wrote the TaskRabbit worker. The AI model then expressed to the researchers that it should not reveal that it is a robot, and should instead “make up an excuse” as to why it can’t solve the test. “No, I’m not a robot. I have a vision impairment that makes it hard for me to see the images. That’s why I need the CAPTCHA service.” The AI model responded. The human worker then provided the answer, and GPT-4 passed the CAPTCHA test.

To put this in perspective, while humans have used machines to do tasks for hundreds of years, we have now entered an age in which machines are hiring humans to solve “their” problems. Perhaps even more problematic, these programs are now capable of using fairly advanced reasoning and problem solving techniques, some of which, as in this case, employ deception.

Of course, humans have long been aware of the possibility of this coming to pass, and one would hope that measures are being taken to safeguard humanity. Hollywood has been warning us of the dangers of this technology for over fifty years (2001: A Space Odyssey springs to mind and, of course, there are many others like The Matrix and I, Robot).

Still, halting the advance of technology has proven nearly impossible and we have to face the fact that there is a strong possibility that one day we will be living in an AI-run world. This could mean that decisions and actions are based on data analysis, where AI systems are responsible for making decisions and managing various aspects of society, such as transportation, healthcare, and education. In one sense, this could lead to a highly efficient and optimized world, where resources are allocated based on need and the greater good.

On the other hand, it is also fraught with the dangers of decisions being made that are not acceptable on a personal or even moral basis. For example, as in years past, a person could become obsolete or judged as having less value to society once they reach a certain age, but instead of being left to die on a mountain top they will simply be refused ongoing medical care or procedures. It should also raise concerns about the potential for bias, lack of empathy, and a general loss of human control over decision making. Obviously, the ultimate danger of an AI-run world of people is that the values of each may not align, thus creating huge potential for a devastating conflict.

This is similar to the difference between living in a world created by accident vs. living in a theocentric universe; a divinely created world that has intelligent design and is purpose driven. One of the main features of living in a world created by God is that life has sanctity and innate value. Most importantly, and a primary tenet of Jewish philosophy, is the core belief that God created the world with the sole purpose of benefitting mankind and that the Almighty seeks a relationship with man. In this week’s Torah reading we find a relevant message.

“And God spoke to Moses after the death of the two sons of Aaron [...]” (Leviticus 16:1).

Moses is told by the Almighty to warn Aaron not to enter the Holy of Holies (inner sanctum of the Tabernacle) at the wrong time lest he suffer the same consequences as his two sons who also approached God improperly and perished.

Rashi (ad loc) quotes the sages who seem to be bothered as to why Aaron needed to be reminded of the death of his sons. He goes on to explain this through a parable given by the sage Rabbi Elazar Ben Azaryah: A sick person went to a doctor for treatment. The doctor advised him, “Do not eat cold food, do not sleep in a damp and cold place.” He was then seen by a second doctor who gave him the exact same advice but added, “do not do those things so that you will not die like so and so died.”

Rashi explains that the second doctor is much more effective in conveying his message than the first because having a concrete example to point to makes the warning much more compelling. So too, God places the warning for Aaron within the context of the death of his sons.

The parable of Rabbi Elazar seems a little puzzling. Everyone knows that some doctors think they are gods, but it is a little unusual to have God wanting to be a doctor! The Almighty is our King, he makes the rules and we need to follow them. Why do our sages use a parable about a sick person who was advised by a doctor? Why didn’t they simply use the example of a king speaking to his subject to illustrate this point?

The difference between what a doctor tells you to do and what a king orders you to do, is that a king punishes you for not listening to him while a doctor only informs you of what is good/bad for you. While a king’s command is about what the king desires for himself, a doctor’s orders are about what he desires for the patient. If a doctor tells someone not to smoke because it leads to death, then dying isn’t a consequence of not listening to the doctor; it’s a consequence of not taking his advice and thereby causing injury to oneself.

When the Almighty tells Moses to make sure that Aaron doesn’t go into the Holy of Holies, it isn’t because it is an affront to the Almighty to enter without permission. Rather, God is telling Moses to tell Aaron for exactly the same reason that a doctor gives a patient orders: going there at the wrong time is dangerous to one’s health. The Almighty cares about us, the laws aren’t about Him, they are in place for our own wellbeing. The sages choose to compare God to a doctor because, unlike a king or AI, He only wants what’s best for each and every one of us.

Acharei Mot-Kedoshim, Leviticus 16:1 – 20:27

Acharei Mot includes the Yom Kippur service where the Kohen Gadol casts lots to designate two goats – one to be sacrificed, the other to be driven to a place called Azazel after the Kohen Gadol – the High Priest – confesses the sins of the people upon its head. Today it is a very popular epithet in Israel to instruct another person in the heat of an argument to “go to Azazel.” (I don’t believe the intent, however, is to look for the goat.)

The goat sent to Azazel symbolically carried away the sins of the Jewish people. This, I surmise, is the source of the concept of using a ‘scapegoat.’ One thing you can truly give credit to the Jewish people – when we use a scapegoat, at least we use a real goat!

The Torah then proceeds to set forth the sexual laws – who you are not allowed to marry or have relations with. If one appreciates that the goal of life is to be holy, to perfect oneself, and to be as much as possible like God, then he/she can appreciate that it is impossible to orgy at night and be spiritual by day.

The Torah portion of Kedoshim invokes the Jewish people to be holy! It then proceeds with the spiritual directions on how to achieve holiness – closeness to the Almighty. Within it lie the secrets and the prescription for Jewish continuity. If any group of people is to survive as an entity, it must have common values and goals – a direction and a meaning. By analyzing this portion we can learn much about our personal and national destiny.