Aipedia: Landscape architects

AI planting plans and species selection

Aipedia, by the Apex broking team · Last reviewed 2026-10-30

AI tools can suggest plant lists and schedules, but the landscape architect remains responsible for choosing species that suit the site, the users and nearby structures. Planting failures are slow to appear and expensive to put right.

Part of: AI in Landscape architects

In short

Generative AI chatbots and planting software can produce plant lists, densities and schedules in seconds. They often ignore local soils, exposure and climate, can include species that are toxic, invasive or aggressive near buildings and drains, and give no reason a client could test. Under LI Rule 6 the designer must be competent to check every choice. When planting fails or causes damage, the client may claim against the designer, and PI usually responds to negligent design, subject to the wording and to whether the cause was design, workmanship or maintenance.

How AI is used in planting design

Practices use AI to draft plant palettes from a brief ("low maintenance, pollinator friendly, coastal"), to convert a palette into a schedule with sizes, densities and quantities, and to write planting notes and establishment requirements. Some design platforms now suggest species automatically as areas are drawn. The tools are good at producing plausible lists. They are poor at knowing that a site has compacted clay subsoil, salt spray, a shaded north-facing courtyard, a children's nursery next door or a shallow-founded extension within a few metres.

The duties that apply

The LI Code of Practice has no planting-specific rule, but its general duties apply to every schedule:

Rule 6: only provide services you are competent to deliver, so each AI-suggested species must be checked by someone with the horticultural knowledge to judge it;
R6g: seek advice where a question is beyond your competence, for example soil science or tree root effects near structures;
R4g(c): a written agreement setting scope, responsibilities and limitations, including who is responsible for soil investigation and maintenance;
R11g(a): honesty and transparency about how the scheme was developed.

The American Society of Landscape Architects' December 2025 AI policy puts it simply: the landscape architect must maintain responsibility for all deliverables, including a schedule a tool drafted.

What can go wrong

The common failure modes with unchecked AI lists are predictable:

species unsuited to the soil, drainage, exposure or regional climate, leading to failed establishment;
plants with toxic berries, leaves or sap specified in play areas, schools or care settings;
species regarded as invasive, or aggressive self-seeders, specified near watercourses or semi-natural habitats;
vigorous trees and shrubs placed close to foundations, drains, boundary walls or services;
schedules with wrong botanical names, cultivars that do not exist or quantities that do not match the drawings.

Illustrative scenario. A practice uses an AI tool to draft the planting schedule for a residential courtyard scheme. The tool includes a vigorous tree species close to a lightweight podium edge and a shrub with toxic berries beside a toddler play space. Two years after completion, the management company reports cracked paving and complaints from parents, and the developer claims the cost of removal, replanting and repairs from the practice.

How PI usually responds

A claim that a planting design was negligent is a professional services claim, and PI usually responds, subject to the terms. Two questions often decide the outcome. First, causation: failed establishment is frequently caused by poor workmanship, poor stock or lack of watering and maintenance rather than design, and the designer's records of what was specified and advised are central. Second, the standard of the obligation: PI generally covers reasonable skill and care, and an appointment that promises a planting scheme will be fit for purpose may go beyond what the policy covers. See fitness for purpose versus reasonable skill and care and professional indemnity insurance for landscape architects.

Property damage from roots can also involve public liability or the contractor's insurance, depending on how it arose.

Practical controls

Before an AI-assisted schedule is issued, a practice can work through these checks:

test every species against site soil data, exposure and the intended management regime;
screen for toxicity where children, animals or vulnerable users are present;
screen for invasive or aggressive species, especially near water and habitats;
check mature size and root behaviour against distances to structures and services, coordinating with tree survey constraints;
verify botanical names and quantities against the drawings;
carry the checked schedule into the specification and maintenance plan.

How Apex helps

Apex arranges PI for landscape architects and planting designers. A broker can review whether your appointments take on fitness for purpose obligations your policy may not cover, and how AI-assisted design is treated. See the landscape architecture hub and PI and AI errors.

Sources

Frequently asked

Can I rely on an AI tool to choose plants?

Not without checking. LI Rule 6 requires competence to deliver the service, so the designer must be able to judge each species against the site's soil, climate, users and nearby structures. AI lists are a starting point, not a specification.

Who pays if planting fails after completion?

It depends on the cause. If the design was negligent, the client may claim against the designer and PI usually responds, subject to the wording. If the failure came from poor stock, workmanship or maintenance, responsibility usually lies with the contractor or the party responsible for aftercare.

Does PI cover tree root damage caused by my planting design?

A claim that the design negligently placed a species too close to a structure is usually a professional negligence claim for PI, subject to the terms. Some damage claims may also involve public liability or other policies, so notifying promptly and letting a broker check the position is sensible.

Check your PI covers how you use AI

Tell us about your firm and the work you do. A broker will review how your current PI treats AI-assisted work, or arrange new cover. No fee for a quote. Or call 0117 325 0027.

Start a PI quote Request a callback

Apex Insurance Brokers Limited is authorised and regulated by the Financial Conduct Authority. Registered in England and Wales, company number 07014570. This page is general information about how AI is used in the profession and how insurance may respond. It is not legal, regulatory or insurance advice on your circumstances, and it does not guarantee that cover will be available or on what terms.