We are approaching a historic juncture. The miracle of AI superintelligence hurtles towards us from one direction. The turmoil of the resulting societal disruption threatens us from the other.
What if we could redeploy the enormous capabilities of an artificial intelligence as a check against the challenge to our social structures?
This is the idea driving the Solomon Project: That the predictable conflicts and chaos brought about by advances in AI might be addressable by those same advanced capabilities. Job destruction, intellectual property, income distribution, unequal access to justice, climate change, property rights—the list of foreseeable flashpoints seems almost unlimited.
At the same time, the startlingly rapid advances in AI frontier models are expanding the possibilities for AI as a tool for social engineering—not through top-down regulation, but through compromise and consensus among stakeholders.
This is the mandate of Solomon. These are the advancing capabilities that will power it:
Advancing insight into human behavior.
Solomon will interpret psychological and emotional states through voice, video, and text. Sensing when a user is trusting or mistrustful, eager for resolution or angling for advantage. This isn’t just sentiment analysis. It’s precisely the intuition that guides an experienced mediator.
Advancing ability to imagine solutions.
Advanced inference will allow Solomon to propose creative, “outside-the-box” terms of agreement that human negotiators might never imagine—solutions that satisfy complex underlying interests rather than stated positions.
Advancing skillsets in communication.
Solomon adapts its language to each party’s communication style and emotional state. Solomon understands, reflects, and validates to soften barriers to agreement. De-escalation isn’t a feature; it’s woven into every interaction.
Advancing ethical understanding.
Solomon is built around mission-driven impartiality—recognizing power imbalances, flagging coercion, and optimizing for fair outcomes rather than fast ones.
Advancing agentic research capabilities.
Solomon acts as a research agent—retrieving financial data, legal codes, precedents, and jurisdiction-specific rules necessary for informed decision-making.
Advancing critical memory windows.
Solomon remembers what matters: prior agreements, stated concerns, evolving positions. It holds the thread so parties don’t have to repeat themselves.
Advancing ability to coordinate complexity.
In multi-party disputes, Solomon tracks issues across stakeholders, maintains consistency, and surfaces interdependencies that might otherwise derail agreement.
Why Now?
These advances may seem hopelessly futuristic. But they are coming—not in some speculative far-off future, but just down the road. Large language models are crossing the threshold from impressive to employable. The public is waking up to the dramatic changes ahead, just as the dream of General Artificial Intelligence approaches reality.
At the crossroads of history, a car crash is possible. Yes, there is still time, although no one can predict with certainty how much. So the Solomon Project is posing the question now: What if there was a system designed from the ground up to meet this moment?
