Hard-Code G-d Into AI

Please see my new article in The Times of Israel called "Hard-Code G-d Into AI."

Artificial intelligence is often framed as a problem of obedience: can we keep machines under human control? But human commands are not always moral, and an AI perfectly aligned with a terrorist, authoritarian state, corrupt corporation or reckless commander would not be safe—it would simply be efficient at doing wrong. This article argues that AI must be built not only with technical guardrails, but around a moral reality higher than machine, market and human command: the sanctity of life, the dignity of every person created b’tzelem Elokim, truth and the duty to protect the vulnerable. That standard does not demand weakness or unilateral disarmament; it calls for AI strong enough to help free societies defeat terrorists, protect civilians and defend themselves, while never confusing power, permission or technological capability with moral authority.

(AI generated image) 

Hard-Code G-d Into AI