Turn a broad belief, strategy or claim into a checked one: say the decision it would change, list every limit it assumes and where each came from, name what would show each limit is wrong, look for evidence far outside the original field, revise both the belief and the question, propose the smallest real test that could overturn it, then run the same questioning on the method itself and repeat at every level it applies to.
Nothing: your AI starts it with your words.
Connect the AI you already use to Trillion once. Then ask in your own words; your AI finds this workflow by what it does and runs it.
Say to your AI: “Use Trillion to turn a broad belief, strategy or claim into a checked one: say the decision it would change, list every limit it assumes and where each came from, name what would show each limit is wrong, look for evidence far outside the original field, revise both the belief and the question, propose the smallest real test that could overturn it, then run the same questioning on the method itself and repeat at every level it applies to”
What your AI calls: start_workflow {"workflow_id":"workflow:question-the-belief-until-it-holds"}