{
  "id": 932465,
  "title": "Open Discovery Challenge: How to Build a Verifiable Judge for AI-Designed Malaria Drugs",
  "url": "https://urgent.news/2026/08/15/open-discovery-challenge-how-to-build-a-verifiable-judge-for-ai",
  "topic": "ai",
  "section": "AI",
  "published": "2026-08-15T02:03:32.000Z",
  "source": {
    "name": "Dev.to",
    "slug": "dev-to",
    "url": "https://dev.to/ai_openfree_b23025ef075cf/open-discovery-challenge-how-to-build-a-verifiable-judge-for-ai-designed-malaria-drugs-430g"
  },
  "original_language": "en",
  "account": "The Open Discovery Challenge aims to create a verifiable judge for AI-designed malaria drugs. The challenge focuses on a malaria parasite enzyme called PfDHODH. A successful drug must inhibit the parasite enzyme without affecting the human counterpart, survive in the environment, and cross cell membranes. The challenge provides a scoring system with six axes, including whole-cell activity, binding, selectivity, ADMET profile, novelty, and synthesis. Validation of the scoring system revealed several issues, such as toxicity gate errors, size bias in molecular weight, and silent errors in novelty scoring. These defects highlight the difficulty of building a fair scientific judge for AI-generated molecules.",
  "summary": "Open Discovery Challenge: How to Build a Verifiable Judge for AI-Designed Malaria Drugs Generative models can propose thousands of plausible molecules in a day. The harder question is no longer whether an AI can draw a molecule. It is whether anyone can tell if that molecule is potent, selective, safe enough to investigate, and possible to synthesize. That is the premise of the Open Discovery…",
  "key_points": [
    "Open Discovery Challenge aims to create a verifiable judge for AI-designed malaria drugs",
    "Successful drug must inhibit PfDHODH enzyme without affecting human counterpart",
    "Scoring system validation revealed toxicity gate errors, size bias, and novelty scoring issues"
  ],
  "editors_take": "The challenge highlights the complexity of creating a fair and reliable evaluation system for AI-designed drugs, revealing issues such as toxicity and bias errors in current scoring methods.",
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}