The thought, architecture, and practice behind the Self-Alignment Framework — the latest below, or browse by topic.
The Five Faculties: A Tour of SAFi’s Cognitive Architecture
Read the article →: The Five Faculties: A Tour of SAFi’s Cognitive ArchitectureMost attempts at AI governance treat alignment as a prompt-level concern. You write a system message, hope the model follows it, and accept that any sufficiently creative attacker can talk the model into ignoring it. The Self-Alignment Framework Interface (SAFi)…
More articles
SAFi Explained: Values
Read the article →: SAFi Explained: ValuesIn the Self-Alignment Framework Interface (SAFi), the faculties of Intellect, Will, Conscience, and Spirit are the fixed, repeatable process, while the Values are the…
SAFi Explained: The Intellect
Read the article →: SAFi Explained: The IntellectThe Intellect is the generative component in SAFi. It is the reasoning faculty and creative force, tasked with producing the initial draft of any…
SAFi Explained: The Will
Read the article →: SAFi Explained: The WillIn the last article, we explored the Intellect, SAFi’s creative engine that generates a draft answer (a_t) and a reflection (r_t). But a purely…
SAFi Explained: The Conscience
Read the article →: SAFi Explained: The ConscienceIn the last article, we saw how the Will acts as a strict gatekeeper, ensuring no response violates the foundational rules of its ethical…
SAFi Explained: The Spirit
Read the article →: SAFi Explained: The SpiritIn SAFi the Intellect proposes, The Will approves, the Conscience audit, and the Spirit integrates. Unlike the other faculties that rely on a LLM…
How SAF Was Born
Read the article →: How SAF Was BornThe Self-Alignment Framework (SAF) did not begin in a lab or a classroom. It began in the quiet space of my own reflection. I…
