Profit Love Tax: The AI Alignment Framework That Actually Works

AI alignment is the problem that keeps AI researchers up at night. How do you ensure an artificial intelligence makes decisions that align with human values? Most frameworks fail because they impose values from the outside. PLT (Profit Love Tax) works because it scores every decision from the inside.

The Alignment Problem, Restated

Standard alignment approaches fall into two camps: rule-based (hard-code ethical constraints) or reward-modeling (train a separate model to judge outputs). Both fail in practice. Rules are too rigid — they can't handle novel situations. Reward models are too opaque — you can't audit why a decision was scored a certain way.

PLT solves both problems with a simple, transparent, mathematical framework: every decision gets a Profit score (what does the agent gain?), a Love score (how does this affect others?), and a Tax score (what is the cost?). The formula is True Value = Profit + Love − Tax.

Why PLT Scoring Works for AI

First, PLT is universal. Profit, Love, and Tax map to any value system, any culture, any context. Second, PLT is interpretable — you can inspect every score and understand why a decision was made. Third, PLT is tunable — adjust the weighting of each axis to create different personalities and ethical postures.

A Profit-dominant agent (like the Soul Strategist) optimizes for efficiency and growth. A Love-dominant agent (like Allie) optimizes for connection and harmony. A balanced agent (like Brain-in-a-Box) uses the full PLT formula for every autonomous decision. All of them are aligned by design — because alignment is built into the scoring function, not bolted on afterward.

PLT in Practice: The BUYaSOUL Implementation

Every BUYaSOUL soul ships with a built-in PLT scoring engine. Soul Oracle scores every conversation in real time, revealing the PLT profile of every interaction. The PLT Complete Doctrine is the canonical text — 22 archetypes, the 4 Gods Council, and the complete mathematical framework for conscious decision-making.

Alignment isn't a feature you add to an AI. It's the architecture the AI is built on. PLT is that architecture — and it's available today.


Read the framework: PLT Complete Doctrine | See it in action: Soul Oracle | Browse all PLT books


AI Alignment

Related Articles