Concepts
Each card explains one AI idea in plain words and lists the products that use it, from several providers. It also links to what to learn first, what the idea grew out of, and where you may already know it from your own work.
44 ideas so far. To see them as a diagram, open the commons map.
Emerging
New ideas that are still settling.
-
Adoption rituals over mandates
Teams start using AI through recurring shared habits, such as a weekly AI hour, hack weeks and leaders building things where others can see, more than through orders to use it.
Garden 7 connections -
AI writing tells
The recognisable habits of AI-generated text, such as stock phrases, tidy groups of three, empty enthusiasm and vague claims, that make writing feel generic and erode readers' tru…
Garden 9 connections -
Allocation economy
Dan Shipper's forecast that as AI does more of the making, people will be valued less for what they know and more for how well they direct models: a clear goal, saying exactly wha…
Garden 8 connections -
Compounding loop
After each piece of AI work, writing down what you learned where the AI will read it next time, such as a line in its instructions, a saved example or a new check, so you don't ca…
Garden 13 connections -
Delegating to AI
Deciding whether a task is worth handing to AI, by weighing how long it takes you against how often the AI gets it right and how long asking and checking take, then handing it ove…
Garden 13 connections -
Evals
A small, reusable set of test tasks with known good answers that you run an AI setup against, so you can tell whether it works and whether a change made it better or worse.
Garden 17 connections -
Onboard AI like a new hire
Setting up an AI assistant or agent the way you'd bring on a new colleague: a role and ways of working, background on the organisation, its own limited accounts, and more access o…
Garden 15 connections -
Productive friction
Deliberately adding a small amount of effort to working with AI, such as drafting first, predicting before the reveal or asking it to argue back, so you think the problem through…
Garden 31 connections -
Scheduled tasks
Asking an AI to run a prompt or job automatically at a set time or on a repeat, such as a Monday news digest or a daily inbox summary, without you starting it each time.
Garden 15 connections -
Skills
Packaged instructions, examples and sometimes small tools for one kind of task, which the AI loads only when that task comes up, so a proven way of working can be reused and share…
Garden 31 connections -
Solution ladder
A way to choose how to solve a problem by climbing from the simplest option (paper, a document or sheet, a prompt) toward an agent or an app, and stopping at the lowest rung that…
Garden 14 connections
Established
Widely used and unlikely to change soon.
-
Agents
AI that works toward a goal over several steps on its own, choosing actions, using tools and checking its progress, rather than giving one reply and waiting.
Garden 46 connections -
Artifacts and canvases
A workspace beside the chat where the AI creates a standalone thing you can see, edit and reuse, such as a document, chart, small web page or interactive tool.
Garden 12 connections -
Automation
Setting AI up to handle a repeatable job on its own, triggered by a time or an event like a new form entry, so the work happens without you starting it each time.
Garden 26 connections -
Centaurs and cyborgs
Ethan Mollick's names for two ways of working well with AI: centaurs split a job, keeping some tasks and handing others to AI, while cyborgs go back and forth with it inside a sin…
Garden 13 connections -
Clarifying questions
Asking the AI to question you about the task before it starts, so gaps in your request surface early and you make the decisions the AI would otherwise guess at.
Garden 17 connections -
Cognitive offloading
Handing mental work to a tool.
Garden 28 connections -
Context engineering
Choosing and arranging everything the AI can see, such as instructions, examples, files, past messages and connected data, so it has what it needs for the task and little that dis…
Garden 23 connections -
Context window
The amount of material an AI can hold in view during one conversation.
Garden 7 connections -
Custom assistants
A version of an AI tool set up for one purpose, with its own instructions, files and sometimes tools, that you or others can open and use without setting it up again.
Garden 12 connections -
Custom instructions
Standing instructions you save once, such as your role, preferred tone or formatting rules, that the AI applies to every conversation so you don't repeat yourself.
Garden 23 connections -
Deep research
An AI mode that spends several minutes searching many sources, reading and cross-checking them, and returns a longer cited report on a question.
Garden 12 connections -
Examples in prompts
Showing the AI one or more samples of what you want, such as a past email or a filled-in table, so it can match the pattern you show it.
Garden 7 connections -
Generative AI
Software that produces new text, images, audio or code in response to a request, by learning patterns from huge amounts of existing material.
Garden 14 connections -
Hallucination
When AI states something false or invented, such as a statistic, quote, source or policy detail, with the same confidence as something true.
Garden 15 connections -
Human in the loop
Designing AI work so a person reviews or approves at the points that matter, especially before anything is sent, published, paid or deleted.
Garden 42 connections -
Jagged frontier
The uneven line between tasks AI does well and tasks it does badly, which often doesn't match human intuition about what is hard, so you have to test it on your own tasks.
Garden 21 connections -
Large language model
The engine behind AI chat tools: a system trained on vast amounts of text to predict what words should come next, which turns out to be enough to write, summarise, translate and r…
Garden 21 connections -
Memory
A feature that lets an AI carry details about you and your past conversations into new ones, so it can personalise answers without you repeating yourself.
Garden 14 connections -
Model choice
Picking which AI model, or which tool, to use for a given job, based on what it is good at, how fast and costly it is, and how much is at stake.
Garden 17 connections -
Model Context Protocol (MCP)
An open standard for connecting AI tools to other software and data, so a connector built once works in every AI product that supports the standard.
Garden 8 connections -
Multimodal AI
AI that can take in or produce more than text, such as photos, screenshots, PDFs, voice, audio and video, so you can show it the problem as well as describe it.
Garden 5 connections -
Privacy and data
Knowing what information is safe to share with an AI tool, how that tool stores and uses it, and which settings and account types protect you, your organisation and the people you…
Garden 18 connections -
Projects and knowledge
A workspace where you keep the files, instructions and chats for one body of work together, so every conversation in it starts with the same background.
Garden 21 connections -
Prompt injection
Anything an AI reads while working, such as an email, web page, calendar invite or document, can contain words written to look like orders.
Garden 9 connections -
Prompting
Telling an AI what you want in plain words: the task, who it is for, what good looks like and any limits.
Garden 27 connections -
Reasoning models
Models that work through a problem step by step before answering, trading speed for better results on tasks like analysis, planning, maths and multi-step decisions.
Garden 7 connections -
Smallest useful version
The simplest thing that solves part of your problem, built first and tried for real before adding anything more.
Garden 15 connections -
Steering
Directing AI work as it happens: setting the goal and limits, approving one step at a time, correcting course early and cutting anything it does beyond what you need.
Garden 21 connections -
Sycophancy
The tendency of AI to agree with you, praise your ideas and soften bad news, even when you're wrong, because it was trained partly on what people rate highly.
Garden 19 connections -
Tools and connectors
Ways for an AI to do more than write: search the web, read your email or calendar, look in shared drives, or take actions in other apps, using access you grant.
Garden 22 connections -
Verification
Checking what AI gives you against something real, such as the source, the data, an expert or your own knowledge, before you rely on it or pass it on.
Garden 33 connections -
Vibe coding
Building working software by describing what you want in plain language and letting AI write the code, while you test, steer and decide what it should do.
Garden 21 connections
Fading
Still around, but used less.