Dear OpenAI team,
I am not a programmer, AI researcher, or government official. I am simply a person living through a war who keeps asking one question: with everything modern AI can already do, why are we still relying almost entirely on humans who are unable or unwilling to find new ways toward peace?
I have an idea I believe is worth serious consideration.
What if OpenAI created a dedicated Multi-Agent Peace and De-Escalation System?
Instead of one AI giving an answer, multiple independent AI agents could work together:
- one analyzes verified facts and intelligence;
- another examines international law and diplomatic constraints;
- another models economic consequences;
- another represents the interests and concerns of each side;
- another actively challenges the assumptions and proposed solutions;
- another searches for compromises and previously overlooked options;
- another stress-tests every proposed scenario for risks and unintended consequences.
The important part is that the AI would not make political decisions or replace governments.
Its role would be to expand the range of possible peaceful solutions and show decision-makers:
“What are the options you may not have considered? What happens if you take this step? What could make the other side accept it? What risks are hidden in this scenario? What alternative paths could reduce violence?”
Governments could have their own secure version of such a system, while independent AI agents could cross-check one another so that no single model becomes the sole source of truth.
I believe this could become something like an AI Peace Laboratory — a place where the world’s most capable AI systems continuously work on conflict prevention, negotiations, de-escalation and reconstruction scenarios.
And perhaps this should not be available only to governments and large corporations. A public version could allow researchers, journalists, humanitarian organizations and ordinary people to explore verified information and possible peaceful scenarios as well.
I am writing this because I cannot build such a system myself. I don’t have the technical resources or money to create a multi-agent AI network.
But OpenAI does.
If humanity is going to build increasingly powerful artificial intelligence, perhaps one of its most important purposes should be helping humans find ways not to destroy each other.
I am asking you to please forward this idea to the people working on AI research, multi-agent systems, safety, international cooperation or conflict-resolution applications.
Maybe this idea already exists somewhere inside your organization. If it does, I would still like to add one simple thought from a person living through the consequences of war:
Please don’t wait until politicians discover how to use AI for war more effectively. Build systems that help them discover how to use AI for peace.
Sincerely,
Ilona