Aiciety maps the big picture: artificial intelligence woven into the structures of work, power, knowledge, and law. But that weaving never happens in the abstract. It happens through millions of small, ordinary acts — a person typing a request to a model and reading what comes back. The prompt is the contact surface, the actual point where human and machine meet. PromptGarden is a project about that surface.
A companion site to aiciety, PromptGarden (tagline: sow smart prompts, reap fresh results) is a growing collection of prompts for getting more out of AI systems — for crafting sharper questions, generating images, creative writing, strategic thinking, and everyday utility. Some entries are playful: a text adventure that drops you inside Kafka’s The Trial, or a historical scene summoned from nothing but a set of coordinates and a timestamp. Some are practical, like handling routine PDF tasks by simply asking a model instead of reaching for specialized software.
And one strand is quietly serious — which is where it connects to aiciety most directly.
A recurring theme in the garden is making AI reveal how it knows what it claims to know. The “Four Doors” prompt asks a model to say whether a given answer rests on knowledge, on inference, on uncertainty, or on a policy constraint — four very different situations that otherwise tend to arrive wearing the same confident prose. The “Confidence Map” asks the model to tag each claim as known, inferred, or guessed, and then to name the single shakiest assumption holding the answer together. These are not party tricks. They are a form of literacy.
That literacy is exactly what aiciety cares about. The worry running through this whole site is that as AI shifts from a tool we use to an infrastructure we live inside, the human risks sliding from informed agent to passive recipient — accepting smooth, authoritative output without seeing the seams underneath. Whether that slide happens is decided in large part at the contact surface: in whether people know how to make the machine show its workings rather than just its conclusions. The prompts that force an AI to expose its confidence and its constraints are small acts of resistance against that drift.
So PromptGarden is best read as the practical counterpart to aiciety’s conceptual map. Where aiciety describes the terrain — the stages of integration, the fields being reshaped — PromptGarden hands you tools for walking that terrain without losing your footing.
The name is well chosen. A garden is tended, not automated. The relationship between people and AI is not something that simply happens to us; it is cultivated, prompt by prompt, choice by choice. PromptGarden is one attempt to cultivate it well, and a natural next stop for aiciety readers who want to move from thinking about the transformation to practicing it.
Visit the project at promptgarden.net.
Leave a Reply