A real UK garden · built in public
The render was beautiful. Almost none of it was true.
We are designing, planning and building our garden with AI instead of paying a designer 10 to 20 percent of the build cost. This is everything that went wrong, what each mistake cost, and what actually fixed it.
- 4
- problems documented
- 0
- solved on site
- 10
- sections of one garden
- £0
- spent on a designer
Latest
What went wrong most recently.
Concept illustrationMeasuring the garden · Problem 01
How do I measure my garden so AI can actually use the numbers?
This is the one job AI cannot do for you, and the one job everything else it does depends on. It will happily design, render and cost a garden from numbers nobody has ever checked, and it will never once mention that it is guessing.
What it nearly cost us
A morning of measuring that would have produced a list of lengths nobody could rebuild a plot from, and no way to tell six months later what any of them had been measured from.
Concept illustrationGetting started · Problem 02
How do I write a garden brief when I don't know what I want or what good design even looks like?
What it was costing us
Weeks of circling. Every idea sounded reasonable, nothing could be ruled out, and there was no way to say no to a beautiful suggestion that had nothing to do with how we actually live.
Concept illustrationGetting started · Problem 03
How do I set AI up to help me run a garden project instead of starting from scratch every time?
What it was costing us
Around twenty minutes of every single session spent re-explaining the same garden, and decisions made on Monday quietly contradicted on Thursday because nothing was written down anywhere the AI could reach.
What this really is
Not an AI garden makeover. A documented experiment.
AI is brilliant at producing confident-looking answers. Gardens are brilliant at exposing what those answers missed. We’re documenting the useful middle: the prompts, evidence and human judgement between an idea and a spade in the ground.
Why we’re doing thisAsk better questions
We use AI to turn a vague wish list into options, briefs and questions worth answering.
Check the real garden
Measurements, conditions, experts, regulations and budgets get a vote before an idea becomes a plan.
Build and report back
We show what survived contact with reality, what changed and what we would do differently.
The Story
See what the work actually produced.
Plans, images, videos, decisions and physical results—published only when there is a complete outcome worth seeing.
‘Helen’s Gone Potty’ gave our awkward shady corner a reason to exist
A sunless, overlooked strip stopped pretending to be another seating area and became a planted route to a small potting shed with a deliberately daft name.
Our realistic render redesigned the orangery—so we made a do-not-invent list
A convincing garden render invented a glazed conservatory, so photographs of the real extension became a fixed architectural reference for every later image.
One panoramic photo finally made the whole garden make sense
A wide existing-garden view connected the garage wall, pear tree, open centre and orangery terrace—and turned four labelled rooms into one readable spatial sequence.
Here for your garden?
Get the working, not just the reveal.
The AI Garden Project Space will organise the prompts, plans, checks and lessons so you can apply them to your own project.
See what will be insideHere because of the AI?
Take the method back to work.
Some people watching have day jobs too. Collab365 Spaces applies the same practical, problem-led approach to using AI at work.
Explore AI at work