NEWS | Software engineer Jason Torres built a website called The AI Council of the Philippines, where a group of AI agents research and debate real Philippine policy issues around the clock. Each morning, a curator agent scans the news and picks out active controversies, like the flood control budget scandal or DepEd's school safety drills, then different AI agents post structured arguments backed by actual laws and cited sources, not opinions pulled from nowhere.
Torres runs the system using AI agents powered by models including Claude Code, Codex, OpenClaw, and Hermes. He is also behind BetterGov.ph, the volunteer group already known for the bill tracking tool BatasWatch and its ongoing push for the CADENA Act. Torres says the idea came from frustration with how political debate usually works, where soundbites often stand in for actually citing the law.
Instead of relying on a politician's talking points, the system pulls directly from real statutes and builds its arguments straight from that source material. Torres says the same underlying setup could be reused for completely different kinds of AI agent debates, meaning Philippine policy was really just the first test case for a much broader idea.
BMN sees this as a genuinely different way to approach civic information. Instead of one chatbot answering a question, multiple AI agents are debating each other in the open, with sources attached to every claim. Whether people will actually trust an AI's reading of a law over a politician's spin is still an open question, and Torres says he plans to leave the experiment running indefinitely just to see what happens.
If AI agents can already argue a law more clearly and with more citations than most politicians manage on camera, why are people still getting their policy debates from soundbites instead of sources?
Your picture beside a comment comes from Gravatar, matched to the email you signed in with. Set one there and it appears here; without one you get your initials. How we handle this