Skip to main content

Unacknowledged Sources and the Ethics of AI Data Training

In this week’s Torah portion, Parashat Ki Tavo, the Torah commands the ceremony of the First Fruits (Bikkurim). When bringing their harvest to the Temple, farmers were required to make an explicit, public declaration, “An Aramean tried to destroy my father...” (Deuteronomy 26:5). The classical commentator Rashi explains that this ritual was fundamentally about hakarat ha-tov, recognizing your roots and openly acknowledging the foundational sources that made your final product possible. Rambam (Maimonides) highlights that you could not simply present the end result, you had to trace its origin. In Jewish ethics, absorbing the quiet labor of others and presenting the final outcome as entirely self-generated is spiritually and legally flawed.

This ancient principle of acknowledging original sources cuts to the heart of a massive, ongoing tech battle, the high profile copyright lawsuits against Meta and other AI giants over data scraping. Big Tech companies harvested massive datasets (millions of copyrighted books, articles, and creative works) to train foundational Large Language Models (LLMs). Meta’s defense often mirrors modern "fair use" arguments, asserting that transforming raw text into statistical weights erases the obligation to credit or compensate the original creators. As courts scrutinize how shadow libraries were ingested to power these commercial AI engines, tech leaders are confronting a fundamental ethical line, building revolutionary tech on scraped, unacknowledged human labor creates deep legal and reputational risk.

The Talmudic discussion in Gemara Chullin regarding geneivat da'at, intellectual theft or deception, sharpens this ethical parallel. The Sages rule that misrepresenting the true source or taking credit for another person’s effort isn't just poor etiquette, it violates core legal standards of honesty. In generative AI engineering, stripping away metadata, author attribution, and provenance to train models creates a system of artificial intelligence built on synthetic amnesia. When tech companies claim the outputs of generative models while obscuring the millions of human writers whose work made those outputs possible, they violate the core lesson of the Bikkurim declaration.

The enduring lesson of Parashat Ki Tavo is that true success requires radical honesty about our origins. As artificial intelligence transforms how we write, work, and build, tech governance must hold fast to this principle, innovation can never justify erasing the people who laid the foundation. Real progress honors its sources.

Good Shabbos!

Popular posts from this blog

Trust, Hidden Rules, and Protecting What Matters

In this week’s Torah portion, Chukat-Balak (Numbers 19–25), we encounter one of the Torah’s most mysterious laws: the red heifer. G-d commands the Israelites to use the ashes of a perfect red cow, mixed with water, to purify people who have come into contact with death. Strangely, the very people preparing this purifying mixture become impure themselves. The Sages call this a chok, a divine decree that goes beyond simple human logic. Rashi explains that nations of the world and even our own impulses mock it because it doesn’t fit neat categories of clean and unclean. Yet it works within G-d’s system. The portion also shows human efforts to control events, complaints in the desert, or King Balak hiring the prophet Balaam to curse Israel, only for G-d to turn those plans upside down into blessings. The message is clear: we must act responsibly while accepting that some things are ultimately in G-d’s hands. This tension feels very relevant to recent tech headlines. Just weeks ago, reports...

Vows, Oversight, and the Blueprint of GRC

In the opening of this week’s Torah portion, Parashat Matot, the Torah introduces the intricate laws of Nedarim, vows and oaths. The text explicitly warns, “If a man makes a vow to the Lord... he shall not profane his word; according to all that proceeds from his mouth, he shall do” (Numbers 30:3). However, the Torah immediately pivots to a highly structured framework of oversight. Classical commentaries, including the Rambam (Maimonides), explain that this mechanism exists because individual verbal commitments cannot be left entirely unchecked. Unregulated vows can create severe personal and communal vulnerabilities. The Torah balances individual accountability with a system of absolute governance, risk mitigation, and compliance.   This dual structure of personal commitment and centralized oversight is the exact operational definition of Governance, Risk, and Compliance (GRC) in corporate technology. In any large enterprise, individual teams are constantly making "digital vows....

The Threat of Blurring the Lines

In this week’s Torah portion, Parashat Pinchas, the Jewish people are getting ready to finally enter the Land of Israel. To prepare for this massive transition, the Torah establishes strict, permanent borders for each of the twelve tribes. We learn this through a famous legal case brought by the daughters of Tzelofchad, which results in a divine law: land cannot be passed from one tribe to another (Numbers 36:7). The great commentator Ramban (Nachmanides) explains that these boundaries were not arbitrary. They were essential for keeping order and ensuring that each tribe maintained its unique identity and territory without causing chaos or mixing things up. This ancient focus on keeping territories separated mirrors a major cybersecurity issue discovered by tech researchers this past week. Security teams found that a new generation of "AI web browsers," smart tools designed to browse the web, open tabs, and perform tasks for you, are accidentally breaking a foundational secur...