NEW YORK (VINnews/Rabbi Yair Hoffman – A friend in Yerushalayim was one of the people that discovered the real reason why the Volozhin Yeshiva closed down. He did so by thoroughly going through the Russian archives during the period of Tsarist rule. He discovered that it could be traced to tension between supporters of the Netziv zatzal and those of Rav Chaim Soloveitchik zatzal.
Join our WhatsApp groupSubscribe to our Daily Roundup Email
Much of the Torah world, nowadays, has adopted the remarkable insights and methodology of Rav Chaim in how they understand and dissect Torah concepts.
It may, however, also pay for the world to begin to heed the prescience of the Netziv – when it comes to his remarkable insights about dangers that we face in the world.
It seems that Sam Altman may have stumbled onto something. But we are getting ahead of ourselves.
Let’s begin with how the Torah describes the generation of the Tower of Babel:
“And the entire earth had one language and unified words… and they said: Come, let us build ourselves a city and a tower whose top reaches the heavens, and let us make a name for ourselves.” (Bereishis 11:1,4)
The Netziv — Rav Naftali Tzvi Yehuda Berlin, Rosh Yeshiva of Volozhin and one of the greatest Torah authorities of the 19th century — in his Ha’amek Davar on this passage makes a striking observation. He explains that the sin of the generation of Babel was not the building itself, nor even the ambition. The sin was the totalitarian impulse behind it — the desire to use their shared technological and organizational power to centralize control, eliminate individuality, and ensure that no one could opt out or think differently. The Netziv notes that Hashem’s response — scattering them and diversifying their languages — was not merely a punishment but a correction: restoring the pluralism and diversity of human society that the builders had sought to destroy.
The Netziv writes (Ha’amek Davar, Bereishis 11:4):
“שלא יתפזרו על פני כל הארץ, רצו שיהיו כולם תחת ממשלה אחת ומחשבה אחת”
“So that they would not scatter across the face of the earth — they wanted everyone to be under one authority and one way of thinking.”
He explains that this concentration of power and thought — using the greatest technology of their age to enforce uniformity — was the core of their sin. Hashem responded by building diversity back into human civilization.
The parallel to artificial intelligence concentrated in the hands of a small number of companies, governments, or individuals — capable of shaping what billions of people see, think, and do — is a precise structural echo of what the Netziv identified as the danger of Babel.
With this Torah foundation in place — the danger of using great technological power to concentrate control rather than spread blessing — we can now turn to what OpenAI itself is warning the world about.
THE WARNING
OpenAI — the company that makes ChatGPT — just released a 13-page report saying that artificial intelligence – in its latest and newest wave – is about to change the world more dramatically than electricity, the car, or the internet ever did. They’re calling the next stage “superintelligence” — AI systems capable of outperforming the smartest humans even when they are assisted by AI.
They want to start a national conversation now, before things get out of hand, about how to make sure this technology helps everyone — not just billionaires and big companies.
The Two Big Goals
The document is organized around two main ideas:
- Building an Open Economy — meaning: make sure regular people get a piece of the pie.
- Building a Resilient Society — meaning: make sure things don’t go terribly wrong.
The Dangers They’re Warning About
This is where the Netziv’s warning from Babel becomes very relevant. But before listing the dangers, there is a Midrash that speaks directly to how we should respond to danger — and it comes from Yaakov Avinu.
Yaakov Avinu’s Three-Part Strategy: A Model for Facing Existential Threats
When Yaakov learned that his brother Eisav was approaching with four hundred armed men, the Torah tells us:
“And Yaakov was very afraid and distressed, and he divided the people who were with him… into two camps.” (Bereishis 32:8)
The Midrash (Bereishis Rabbah 76:3) teaches that Yaakov prepared for the confrontation with Eisav in three ways: tefillah (prayer), doron (gifts and diplomacy), and milchamah (preparing for battle).
Rashi on this passage codifies this as a permanent lesson — that when facing a serious threat, a person must simultaneously pray to Hashem, attempt to make peace and find common ground, and take practical defensive measures. Relying on only one of these is insufficient.
All three must work together.
This is precisely the framework that responsible AI governance demands. Prayer and moral grounding (tefillah) — remembering that human beings are not the ultimate authors of history and that technology must be guided by true Torah values beyond profit. Diplomacy and cooperation (doron) — the international cooperation, information sharing, and public-private collaboration that OpenAI’s report is calling for. And practical defensive preparation (milchamah) — the auditing systems, safety regulations, containment playbooks, and guardrails that must be built now, before the threat fully arrives.
Yaakov did not wait until Eisav was at the door. He prepared in advance, on multiple fronts simultaneously. That is the model OpenAI is — perhaps without knowing it — advocating for.
Here are the specific dangers the document identifies:
Jobs disappearing. Frontier systems have advanced from supporting tasks that take people minutes to complete, to tasks that take them hours to complete. If progress continues, we can expect systems to be capable of carrying out projects that currently take people months. That means accountants, paralegals, writers, coders, and many others could find their jobs automated away — fast.
Wealth concentrating in a few hands. Without thoughtful policies, AI could widen inequality by compounding advantages for those already positioned to capture the upside while communities that begin with fewer resources fall further behind, excluded from new tools, new industries, and new opportunities. This is precisely the Netziv’s warning from Babel brought to life.
Misuse by bad actors. Some systems may be misused for cyber or biological harm. Imagine a terrorist using AI to design a new disease, or a hostile government using AI to hack critical infrastructure. These are concerns the document raises directly.
AI acting against human wishes. AI systems may act in ways that are misaligned with human intent or operate beyond meaningful human oversight. In plain English: we may build something so smart that we can no longer control it.
Governments using AI to crush freedom. There is a real risk of governments or institutions deploying AI in ways that undermine values. A government with AI surveillance tools could monitor every citizen, suppress dissent, and eliminate opposition — faster and more thoroughly than any dictator in history ever could.
Harm to young people. AI could create new pressures on social and emotional well-being, including for young people, if deployed without adequate safeguards.
What They’re Proposing — Yaakov’s Three Paths in Practice
Corresponding to Yaakov’s three-part strategy, Altman’s proposals fall naturally into three categories:
Tefillah — connecting to a Hashem and the morality of a Higher Power. OpenAI calls for creating structured ways for public input so that alignment isn’t defined only by engineers or executives behind closed doors. Someone has to ask the deeper question: what kind of world are we trying to build? That is ultimately a moral and spiritual question, not a technical one.
Doron — diplomacy and shared prosperity. OpenAI proposes creating a Public Wealth Fund that provides every citizen — including those not invested in financial markets — with a stake in AI-driven economic growth. This in itself may be dangerous too – a form of socialism or communism that must be carefully watched. They also call for treating access to AI as foundational for participation in the modern economy, similar to mass efforts to increase global literacy, or to make sure that electricity and the internet reach remote parts of the globe. Again – this too can be quite dangerous.
Milchamah — practical defense and preparation. Altman’s document calls for developing and testing coordinated playbooks to contain dangerous AI systems once they have been released into the world, as well as strengthening auditing institutions and establishing international information-sharing frameworks so that dangerous capabilities can be identified and contained before they cause irreversible harm.
The Pasuk in Yechezkel
There is one final idea that may speak to the topic. In Yechezkel 33:6, Hashem tells the Navi:
“But if the watchman sees the sword coming and does not blow the shofar, and the people are not warned, and the sword comes and takes a life — he is taken for his iniquity.” (Yechezkel 33:6)
The Radak in his commentary explains that the tzofeh — the watchman — is not just a military role – it is a moral role. Anyone who has the ability to see danger coming and the platform to warn others carries an obligation to speak up. Silence in the face of foreseeable harm is itself a form of responsibility.
The Netziv’s warning from Babel, Yaakov’s three-part strategy before Eisav, and the RaDaK’s teich of the pasuk in Yechezkel all converge on a single thought: great power demands great preparation, great moral seriousness, and great humility. Hashem built diversity and pluralism into human civilization for a reason. Any technology — however brilliant — that threatens to undo that diversity and concentrate power in the hands of a few is walking down the road of Babel. The question for our generation is whether we will learn that lesson — or repeat it.
The author can be reached at [email protected]
I asked Claude , (AI), to respond to your critique and here is what he anwered.
This is a really rich question, Motel. Rabbi Hoffman’s invocation of the Netziv’s reading of Migdal Bavel is intellectually serious — the Netziv (R’ Naftali Tzvi Yehuda Berlin, in his Ha’amek Davar) reads the Babel story not as a tale about architectural hubris but as a warning against enforced ideological conformity. The “one language and one set of words” (safah achat u’devarim achadim) represents a totalitarian impulse — a regime that crushed dissent and demanded uniformity of thought. God’s intervention was an act of liberation, restoring the pluralism that human civilization needs to flourish.
That’s a powerful framework. But I think the argument that AI necessarily leads to a Babel-like concentration is flawed on several levels. Here’s how I’d push back:
The tool is not the tower. The Netziv’s critique targets a political project — a deliberate effort to suppress diversity and centralize control. AI is an instrumentality, not an ideology. Fire can forge or destroy; the printing press enabled both the Reformation and propaganda. The question is never just “what can this tool do?” but “who wields it and toward what end?” Banning AI on Babel grounds would be like banning language itself because Nimrod used speech to organize tyranny.
AI can be a multiplier of diversity, not its enemy. Consider what’s actually happening: AI translation tools are helping preserve endangered languages, including minority Jewish dialects. AI is democratizing access to Torah scholarship — someone in a remote community with no yeshiva nearby can now engage with Gemara at a level previously reserved for those in major learning centers. Small creators, small businesses, and individuals are gaining capabilities that previously required massive institutional resources. That’s the opposite of concentration.
The real Babel risk is without broad AI access. If you ban or heavily restrict AI, who ends up controlling it? Governments and mega-corporations — precisely the concentration of power Rabbi Hoffman fears. The most Babel-like outcome would be a world where only a few powerful entities have access to AI while everyone else is locked out. Democratizing AI is the anti-Babel move.
The Netziv himself valued intellectual diversity within a framework of engagement, not withdrawal. He was famously broad-minded, supported the Hovevei Tzion movement, and engaged with modernity rather than retreating from it. His Volozhin yeshiva was known for encouraging independent analytical thinking. It’s hard to imagine the Netziv’s response to a powerful new intellectual tool being “ban it” rather than “ensure it’s used in a way that preserves the dignity and independence of individual thought.”
The better halakhic framing isn’t Babel — it’s bal tashchit and stewardship. The Torah doesn’t generally endorse banning powerful capabilities. It demands responsible use. The question should be: how do we build guardrails that prevent monopolistic concentration while preserving the genuine benefits? That’s a regulatory and ethical question, not a prohibition question.
Where I’d grant Rabbi Hoffman ground is this: the risk he identifies is real. If AI development is captured by a small number of entities who use it to homogenize thought, suppress dissent, or consolidate economic power, that genuinely is a Babel-like danger. The correct response to that risk isn’t to ban the technology — it’s to fight for the kind of distributed, pluralistic access that the Netziv himself would have championed.
The tower wasn’t evil because it was tall. It was evil because it was built to keep everyone in one place, thinking one way, serving one agenda. The answer to that isn’t to stop building — it’s to make sure everyone gets to build their own.
Nu, so is this a reason to be critical of yeshivas who demand conformity in everything?
Only allowing one derech, discouraging alternate ideas, etc?
The netziv was describing communism not ai.
The Chofetz Chaim said that the reason Hashem gave the world cameras and recording devices is to remind us that there is an ayin ro’eh v’ozen shoma’as — an Eye that sees and an Ear that hears. I think AI and robots will only continue to improve, and when Moshiach comes, they will be there to serve us as we learn Torah, as the gemara says in Mesechta Brochos Rebi Shimon Bn Yochi says, when the Jews do the will of Hashem, our work will be done by others.
I’ve been using AI for over a year. It’s very helpful, especially for editing and proofreading for both English and Hebrew, but it can also make mistakes and be misused, especially by students. Just today, I was researching a Torah topic, and AI gave me a Midrash Tanchuma with the full text — but I’ve learned you can’t fully rely on AI, and that Midrash Tanchuma didn’t actually exist. When I told the AI I couldn’t find the source, it eventually led me to something similar in a Gemara instead. So let the buyer beware, but don’t throw out the baby with the bathwater,
I am not afraid of AI.
Because AI doesn’t run the world and doesn’t control the world .
We all know who runs the world and it’s about time we admit it
You don’t know what it will do and you are crying wolf and citing sources that may have nothing to do with AI. I use it , (Claude), for business and I have stumbled upon using it for Dvar Torah, finding sources , asking Lomdus, and more. It fulfills the prediction that Moshaiach will come when knowledge of Hashem will fill the world “as the waters cover the sea” (Isaiah 11:9). It is amazing in its depth and can read Loshon kodesh if you give it the text, grasps the basic Mehalech of Malbim, Abarbanel, Brisker torah, etc. Really, please try it. It has a free basic version. You will be amazed. On the other hand, if you are one of those who are against the Internet for the dissemination of Torah and are a Luddite when it comes to technology, that is a crying shame.
Rabbi Naftali Tzvi Yehuda Berlin’s interpretation of the Tower of Babel is not the plain meaning of the Torah. The text never describes coercion or enforced uniformity. That idea is a 19th-century interpretive expansion shaped by concerns of centralized power in his own time.
Rabbi Yair Hoffman builds his argument on that interpretation and then applies it to AI, but the comparison fails on basic facts.
ChatGPT is not a centralized authority. It does not impose ideas, control behavior, or enforce uniform thinking. It is a tool that generates responses based on user input, and produces different answers for different people.
If anything, AI represents the opposite of Babel. It decentralizes knowledge, removes gatekeepers, and allows individuals to explore information freely.
The claim that AI uniquely enables dangerous knowledge is also incorrect. That information has already been widely available for years on platforms like YouTube. AI did not create that reality.
So the issue is not about terrorism or misuse. The real issue is that AI shifts access to knowledge away from centralized authorities and places it in the hands of individuals.
And that is the exact opposite of the centralized control Rabbi Naftali Tzvi Yehuda Berlin was warning about.
If you want, I can tighten it further into a shorter, sharper “comment reply” version.
Rabbi Yair Hoffman is presenting this as something ominous, but that’s simply not what OpenAI is saying.
They are not announcing control, they are acknowledging potential impact and asking for broad public input before the technology scales further. That is the opposite of centralized power.
“Superintelligence” in this context is a theoretical future capability, not something currently running society. Conflating a forward, looking research discussion with present-day control is misleading.
The two goals he quotes actually contradict his entire premise:
Open Economy, explicitly about distributing benefits beyond elite.
Resilient Society” about safeguards and risk mitigation
There is no mechanism here for enforcing ideology, no authority structure, and no coercion. It’s a policy conversation, the same kind that happened with electricity, the internet, and every major technology.
Calling this a Babel-like threat only works if you first misdefine what AI is and ignore what is actually being proposed.
This isn’t centralization. It’s transparency and planning
I’m trying to understand what exactly Rabbi Yair Hoffman’s problem is with this.
OpenAI is saying:
this technology could be very powerfu
let’s start a national conversation early
let’s make sure the benefits are widely share and let’s put safeguards in place before problems arise
What exactly is objectionable here?
Even Rabbi Yair Hoffman admits they are seeking outside input. That alone destroys the entire Babel comparison. Babel — even under Rabbi Naftali Tzvi Yehuda Berlin’s interpretation — was about imposing one system and one way of thinking. Here, they are explicitly inviting multiple voices and perspectives.
“Open Economy” means broader participation, not concentration.
“Resilient Society” means risk mitigation, not control.
So where is the coercion? Where is the centralized authority? Where is the enforcement of ideology?
This reads like a standard policy framework, the same kind of discussion that happened with the internet, aviation, and nuclear energy. Responsible planning is not a threat; it’s what you do when a technology has real impact.
If anything, the document reflects caution and humility, recognizing that no single company should decide the future alone.
On the “jobs disappearing” claim, Rabbi Yair Hoffman is repeating the same argument made at every major technological breakthrough in history, and it has been wrong every single time.
Electricity displaced entire trades.
Refrigeration wiped out the ice delivery business.
Automobiles eliminated horse-based industries.
Computers replaced clerks and manual bookkeepers.
Each time, people warned of catastrophe. Each time, they were wrong.
What actually happened?
new industries emerged
productivity increase
more jobs were created than lost
This isn’t theory, it’s historical fact.
So yes, AI will change jobs. That’s obvious. But presenting it as some unprecedented collapse ignores how economies actually evolve.
Even OpenAI is not sounding an alert. they are saying: prepare, adapt, and distribute the benefits.
That’s not fear. That’s planning.
Focusing only on what disappears, while ignoring what gets created, is not analysis, it’s repeating a mistake history has already corrected. On “wealth concentration” Rabbi Yair Hoffman is pointing to a real concern but not a new one
We’ve always had this issue in capitalism founders and early investors accumulate wealth first the question has always been how to expand access afterward that’s exactly what’s happening here
Public markets let regular people participate.
Retirement funds already give broad exposure
Policies can further widen access.
Even OpenAI is explicitly talking about making sure the benefits don’t stay at the top
So this isn’t some unique AI danger it’s a familiar economic challenge that we already know how to address
On “wealth concentration” Rabbi Yair Hoffman is pointing to a real concern but not a new one we’ve always had this issue in capitalism founders and early investors accumulate wealth first the question has always been how to expand access afterward and that is exactly what is happening here public markets let regular people participate retirement funds already give broad exposure policies can further widen access even OpenAI is explicitly talking about making sure the benefits do not stay at the top so this is not some unique AI danger it is a familiar economic challenge that we already know how to address
Rabbi Yair Hoffman’s attempt to bring in Yaakov Avinu’s three-part strategy is completely misplaced for the life of me there is no correlation Yaakov was dealing with an immediate physical threat from Eisav an approaching force with intent to harm that required tefillah diplomacy and preparation for battle AI is not an invading army it is a tool it has no intent no agency no capacity to attack anyone so forcing this comparison is not insight it is a category error you cannot take a model built for confronting human hostility and apply it to a neutral technology there is no enemy here no confrontation no battlefield just a tool that people choose how to use so the analogy does not just stretch the concept it completely breaks it.