Write a white paper for an AI system for medication/compound...

Prompt

Write a white paper for an AI system for medication/compound discovery. The concept, as intellectual property, is owned by Alexander P. Scott, a 16-year-old visionary who believes in using generative AI for good (LLMs, less so image, music, and video generation models), instead of waste. Essentially, the system will run 100s-1000s of agents in parallel to discover new medications. The first step is a detailed, user-given goal. The system will then reason about this goal. Then it will perform research using a state of the art, vector indexed library of scientific and medical texts from universities and research hospitals. After performing research, it will begin the '100s-1000s of agents in parallel' approach. After these each of these agents reach an end result, a judge/evaluation model (finetuned from a Kimi model in-house) will consult the research library and choose the best result by scoring them. A third agent will then test the result against an atomic-level physics and chemical sandbox to verify it. If it doesn't work, the process repeats itself. When a result works, a fourth agent will explain the result and provide it in detailed, human-readable information to a randomized network of chemists and doctors internationally. These doctors will provide feedback, which will then be evaluated by a fifth agent. After the fifth agent finishes evaluating it, it will determine if the result is ready or needs further refinement. Return the complete white paper.