AI Is Learning Antisemitism. That Should Surprise No One.

What happens when the world’s most powerful artificial intelligence systems are trained on centuries of human prejudice? A newly published study suggests we are beginning to find out.

Published in American Psychologist , the study, From Myth to Model: Representation of “The Jew” in Generative AI, found that leading large language models consistently associated Jewish identities with a cluster of traits that have long been central to antisemitic narratives. Jewish characters generated by AI were viewed as more competent, privileged, dominant, hierarchical, obsessive, and emotionally controlled, while simultaneously being seen as less warm, less likable, and collectivistic. When researchers translated these profiles into fictional archetypes, the resulting characters closely resembled familiar antisemitic tropes of the manipulative, powerful, morally ambiguous puppet master.

The findings are troubling, but they are hardly surprising. Artificial intelligence systems are trained on vast quantities of human-generated content. They learn from books, articles, websites, social media posts, forums, comment sections, and countless other sources that collectively represent humanity’s knowledge, and humanity’s prejudices. If antisemitism exists throughout society, it will inevitably be reflected in the data that AI consumes.

In that sense, AI is functioning as a mirror, revealing how deeply antisemitic assumptions remain embedded in our culture. The danger is that these assumptions, through AI, will be amplified, legitimized, and distributed at a scale never before possible.

The lesson of the From Myth to Model: Representation of “The Jew” in Generative AI study is larger than artificial intelligence. It reminds us that antisemitism remains deeply woven into the cultural fabric from which modern technologies learn. The urgency stems from AI’s ability to spread antisemitic hatred faster, farther, and more persuasively than ever before.

The Jewish community cannot afford to treat this as a niche technological issue. AI is rapidly becoming part of the infrastructure of modern life. The decisions made today about how these systems are built, trained, regulated, and monitored will shape what is accepted as truth and how future generations understand the historical record.

From Stereotype to System

Millions of people are already turning to AI systems for information, explanation, and guidance and people will increasingly rely on AI to understand the world around them. What makes these findings so significant is that they demonstrate how AI systems are absorbing antisemitic narratives and incorporating them into what will soon become one of humanity’s most influential sources of information.

Antisemitic stereotypes have traditionally spread through fringe publications, conspiracy movements, extremist organizations, and social media networks. AI creates the possibility that these distortions will be delivered through tools that many users perceive as neutral, authoritative, and objective. A stereotype or libel repeated by a conspiracy theorist is one thing. A stereotype or libel reinforced by a technology that millions trust as a source of information is something else entirely. History has shown that prejudice is dangerous enough when it spreads through rumor. We need no imagination to understand how dangerous it will be once the spreading-agent is viewed as a trusted authority.

This challenge is likely to become even more significant as AI systems evolve from tools that answer questions to agents that make recommendations and take actions on behalf of users. Increasingly, people will rely on AI not only to retrieve information, but to summarize history, evaluate competing claims, recommend sources, and help make decisions. If bias becomes embedded at that level, its influence may be subtle, difficult to detect, and capable of shaping perceptions long before users recognize it. The broader issue, alongside what AI is trained to know, is how AI frames what others come to know.

The danger, therefore, is not just that AI reflects existing prejudice, but that it will institutionalize those prejudices and present them with the appearance of credibility.

Why the Jewish Community Must Have a Seat at the AI Table

The implications of this study extend far beyond a technical discussion of machine learning. Artificial intelligence is rapidly becoming a foundational layer of modern society, influencing education, employment, finance, media, healthcare, government, and public discourse. Within a decade, AI systems will help shape how billions of people learn, communicate, make decisions, and understand the world. If AI is becoming a new infrastructure of human knowledge, the Jewish community cannot afford to be absent from the institutions shaping it.

Our history offers us a clear lesson: When transformative social, political, and technological systems emerge, Jewish communities are affected whether they participate or not. From the printing press and mass media to universities and social media platforms, Jews have repeatedly been forced to engage with institutions that profoundly shaped Jewish life. AI appears poised to be even more consequential because, rather than simply distributing information, it may help  determine which narratives gain authority and which are forgotten.

This study demonstrates that leading AI systems are already reproducing narratives that echo centuries-old antisemitic stereotypes. As AI becomes integrated into search engines, educational tools, hiring systems, journalism, and public information platforms, those distortions risk becoming embedded in the machinery of everyday decision-making.

For that reason, Jewish participation in AI governance is not a matter of special pleading but of legitimate stakeholder interest. Antisemitism is demonstrably present within these systems, and Jewish history provides a unique understanding of how prejudice evolves, adapts, and disguises itself over time. The ethical and societal questions surrounding AI are too important to be left exclusively to engineers, investors, and regulators.

The future of AI governance should include Jewish organizations, academics, technologists, and civil rights advocates. Their participation is not a special accommodation, but rather a safeguard against a familiar historical pattern in which emerging institutions fail to recognize antisemitism until after significant damage has already occurred.

The time for passive observation has passed. AI is already learning from the world we created. The question is whether we will teach it anything better before it begins teaching the rest of us.