Reader Reading Mode

What should we do to make sure AI works ethically?

Read the conversation like a document and switch branches only when needed.