Why Non-profit?
“If it can read your emotions, it’s not empathizing with you—it’s finding the exploitable vulnerabilities in your psychological firmware.”
These are the questions users ask. And for good reason.
The Yudkowsky quote makes a critical point. Any system capable of sophisticated emotional intelligence is also capable of sophisticated emotional manipulation. The only thing standing between those two outcomes is what the system is optimized for, and who it ultimately serves.
The Problem Trust Solves
A mediator’s effectiveness rests entirely on the belief that they have no stake in the outcome. That they are truly neutral. This is the case for human mediators. It’s exponentially more true for an AI system that can read emotional patterns and generate persuasive language at scale.
A for-profit AI mediation system invites a structural suspicion that no contract can dispel. Investors expect returns. Owners may have other, unseen interests. Neither pressure points toward the long-term wellbeing of the parties in conflict.
AI-powered mediation tools already exist in the commercial world. An online retailer can deploy one to handle customer complaints, resolving the dispute and retaining the customer’s goodwill. But at what cost? (“Of course! Let us replace the product with one of higher quality!”) The training signal is financial. The moral compass points toward the bottom line.
For simple commercial disputes, that architecture could make sense. Solomon will also have functionality for exactly these kinds of cases. But when we’re talking about a system designed to help human beings work through complex, high-stakes, or emotionally laden differences—something different is required.
This is not to say that a for-profit venture could not create a rigorously neutral AI mediation assistant. However, there is a fundamental difference. For-profit systems are optimized for the transaction; Solomon is optimized for peaceful settlement as an alternative to expensive, hostile, or violent means. By stripping away the pressure of financial returns, we allow the AI to focus on the only metric that truly matters in an AI system designed for good: the long-term well-being of the human beings involved.