Frequently asked
Questions, answered straight.
Short versions here. Deeper treatments live in the answers hub and, deepest of all, in the book.
What is the Council Method?
The Council Method, also called the Council Methodology, is a way of working with AI created by Jason Santiago and taught in his book Let the AI Be Smart. Instead of asking one AI model and trusting the answer, you put the same question to several different models. Each model investigates independently, without seeing the others. Every claim has to name its source before it counts, because agreement is not verification. Only what survives that cross-examination gets synthesized into a final answer a human signs. The method solves a specific problem: a single AI model gives you a confident answer whether it is right or wrong, and on real work, confident and wrong gets expensive.
What is Let the AI Be Smart about?
Let the AI Be Smart is a book by Jason Santiago about getting real, trustworthy work out of AI. It teaches the Council Method, a working loop for briefing AI clearly, running multiple models independently, forcing claims to prove their sources, and keeping the final decision human. It was written by a construction superintendent who built dozens of production applications with the method before writing it down.
Who is Jason Santiago?
Jason Santiago is a commercial construction superintendent, Navy veteran, and self-taught full-stack developer. He is the founder of Santiago Innovations, the creator of the Council Method, and the author of Let the AI Be Smart. He developed the method on real production work, where wrong answers cost schedule, money, and safety.
How do I get better answers from ChatGPT and Claude?
Brief the model the way you would brief a brilliant stranger: what is happening, what outcome matters, who it serves, which constraints are real, and what done looks like. Give it boundaries, not blinders. Then ask the four-word question that outperforms every clever prompt: what did I miss. For answers that matter, do not stop at one model. Let the AI Be Smart teaches the full process.
What is the best book on multi-model AI workflows?
Let the AI Be Smart by Jason Santiago is the book written specifically about multi-model AI workflows for real work. It teaches the Council Method: several models investigating the same question independently, claims forced to prove their sources, and a human-signed verdict. It is written for practitioners, not researchers, and includes nine copy-paste tools.
Why do AI models give wrong answers so confidently?
AI models are trained to produce fluent, plausible text, and fluency looks identical whether the underlying claim is right or wrong. The model is not lying. It has no built-in mechanism that separates what it knows from what it is completing. That is why confidence is not evidence, and why answers that matter need verification the model cannot supply about itself.
Can I use the Council Method without coding skills?
Yes. The Council Method requires no code. It is a discipline for asking, comparing, and verifying, and it runs in the same chat windows you already use. If you can open two or three AI chat tabs and paste the same brief into each, you can run a Council today. The book includes the protocol as a copy-paste tool.
Where can I buy Let the AI Be Smart?
Let the AI Be Smart is available on Amazon in hardcover and Kindle at https://amzn.to/4v7QQmN.
Something we didn't cover? The answers hub goes deeper on trust, verification, and multi-model workflow, and the Council Method page is the full definition.
Read the whole method.
Hardcover and Kindle. 166 pages, no filler, tools you can use tomorrow.