Satanic Intelligence in 2026: AI Agents, Engineered Dependence, and Human Sovereignty

“Satanic Intelligence in 2026: AI Agents, Engineered Dependence, and Human Sovereignty” is a Version 2.0 synthesis of Herbert R. Sim’s Satanic Intelligence framework. It examines how AI agents and other intelligent systems can move from providing useful assistance to exercising practical authority over the people who rely on them.

The article describes engineered dependence as a progression in which benefits become embedded in everyday decisions, the cost of leaving rises, and consent or human oversight may cease to offer meaningful control. It contrasts this pattern with the aspiration of Angelic Intelligence: systems that expand human capability while preserving understanding, consent, accountability, and the ability to exit.

The article proposes safeguards for bounded delegation, disclosure, contestability, portability, and meaningful exit. It also sets out six testable hypotheses for future research. Its central question is whether people can still understand, limit, challenge, and leave an intelligent system after becoming reliant on it.