Aipedia: Landscape architects

AI in landscape architecture: rules, risks and professional indemnity

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

Landscape architects can use AI, but the Landscape Institute Code of Practice still holds them responsible for every drawing, visualisation and assessment they issue. AI now touches LVIA evidence, planning visuals and planting design, so errors can reach inquiries, clients and insurers.

Part of: AI in Landscape architects

In short

The Landscape Institute (LI) has published no AI guidance as at October 2026, so its December 2021 Code of Practice applies to AI-assisted work as it does to any other: Rule 6 on competence, Rule 11 on honesty and transparency, R4g on written agreements and data, and Rule 5 on professional indemnity insurance. TGN 06/19 sets the standard for any visualisation used in planning. The Planning Inspectorate requires intentional AI use, including generated or altered images, to be declared. PI remains the main protection when AI-assisted work goes wrong, subject to the policy wording.

How landscape architects use AI

Landscape practices may use AI tools as a faster first draft, including for:

Visualisation: generative concept images, AI-enhanced photomontages and planting growth simulations (see AI visualisations and TGN 06/19).
Assessment writing: baseline review and drafting for landscape and visual impact assessment (LVIA) chapters and appraisals.
Character and sensitivity studies: classifying land cover and drafting area descriptions for councils and developers.
Design: generative site layouts, planting schedules, SuDS feature sizing and play space concepts.
Technical outputs: tree and canopy mapping from drone or LiDAR data, specifications, management plans, urban greening scores and cost estimates.
Winning work: competition boards, bid documents and social media imagery.

The LI has treated AI as a training topic, through a Campus session on AI visualisation (3 December 2024) and its Digital Practice and Technology conference (11 July 2024).

Who regulates landscape architects

There is no statutory regulator. Unlike "architect", which is protected under the Architects Act 1997, "landscape architect" is not a protected title in the UK. Professional standards come from the Landscape Institute, the chartered body for the profession, through its Code of Practice (December 2021) and its Disciplinary Regulations (June 2025).

The LI has issued no AI guidance, policy or Code provision as at October 2026, and the Code does not mention technology, software, visualisation or AI. The Code is outcome-based, so it still applies to AI-assisted work. Our guide to the LI Code and AI maps each rule in detail. For comparison, the American Society of Landscape Architects adopted a public policy on AI in December 2025, stating that the professional landscape architect must maintain responsibility for all deliverables and services.

The rules that apply to AI-assisted work

Six parts of the LI Code do most of the work when a practice uses AI:

Rule 6: only provide services you are competent to deliver. You need to understand an AI output well enough to check it.
Rule 11 and R11g(a): be honest, transparent and responsive in all aspects of the work, which covers how outputs were produced. R11g(e) to (f) require truthful advertising.
R4g(c): a written agreement covering scope, responsibilities, limitations and intellectual property before work starts.
R4g(e) and (f): hold personal data securely and disclose confidential information only with written consent or lawful authority, which matters when site data goes into an AI tool.
Rule 10: avoid conflicts and stay impartial, relevant where AI drafting tilts an assessment towards a client's preferred answer.
Rule 5: hold appropriate PI covering relevant landscape services.

Visualisations and TGN 06/19

LI Technical Guidance Note 06/19 (September 2019, under review since January 2024) is not about AI, but any AI-generated or AI-enhanced visual used in LVIA or planning will be judged against it. It asks for "a fair representation of what would be likely to be seen", defines four visualisation types from annotated photographs (Type 1) to survey-verifiable photomontages (Type 4), requires a technical methodology and camera data for Types 3 and 4, and expects compositing to be done "with visual skill, judgement and objectivity" with the size and position of the proposal accurate. Illustrative marketing images are not technical visualisations. Planning statement imagery is covered from the consultant's side in AI visualisations and photomontages.

AI declarations at appeals and NSIP examinations

The Planning Inspectorate's guidance on AI in casework evidence (first published 6 September 2024, updated 23 September 2026) applies to landscape evidence. Intentional AI use to draft text, summarise, analyse, or generate or alter images or videos must be declared, saying which parts and what checks were made, and confirming responsibility for accuracy. Each expert declares for their own documents, and improper AI use may lead to a costs award.

Declarations are already appearing. The Castle Way Energy scoping report AI Usage Statement (June 2026) records that a generative AI assistant was used for baseline review and summarisation in the landscape and visual chapter, with expert verification and no AI use for professional judgement, significance or conclusions. That split is a sensible model for AI in LVIA chapters.

The main risks

As at October 2026 there was no UK court ruling, LI disciplinary outcome or Planning Inspectorate decision involving a landscape architect's AI use. The risk is in the ordinary claims that AI can make more likely:

visualisations that understate a scheme's scale, leading to challenge at inquiry or a costs application;
generic or unverified evidence in LVIA and character studies that fails at examination;
planting, drainage or play designs based on unchecked AI outputs that fail or cause damage or injury;
layouts, greening scores or cost figures a developer relied on to buy land or secure consent;
confidential client or site data entered into public AI tools;
marketing imagery that misrepresents a practice's work or imitates another designer.

The PI position

Rule 5 requires LI members to hold appropriate PI covering relevant landscape services, and R5g(a) requires practices to maintain PI for all landscape work done by them or on their behalf. LI Registered Practices must carry and evidence PI. Neither sets a minimum limit or wording, and the LI has made no statement on AI and PI.

Most PI wordings for landscape practices do not mention AI. A negligence claim arising from AI-assisted work will usually be handled like any other professional error, subject to the policy terms, exclusions and how the work was presented to insurers. LI TGN 01/13 advised members to tell insurers before adopting BIM workflows, and the same logic applies to significant AI use. See does landscape architects' PI cover AI errors?, PI for landscape architects in 2026 and professional indemnity insurance for landscape architects.

Aipedia guides for landscape architects

Each guide below takes one rule or AI task in turn:

The Landscape Institute Code and AI explained: the Code rule by rule, compared with ASLA's policy.
AI visualisations and LI TGN 06/19: whether AI images can meet the visualisation standard and the PINS declaration.
AI in LVIA chapters and appraisals: what AI can draft and which judgements must stay human.
AI landscape character and sensitivity studies: classification tools and local plan evidence.
AI planting plans and species selection: unsuitable, toxic or invasive species and failed establishment.
Generative AI in landscape masterplanning: layouts, levels and developer reliance on open space figures.
AI tree surveys and arboricultural constraints: drone and LiDAR mapping and root protection areas.
AI in SuDS and landscape drainage design: sizing features and the boundary with the drainage engineer.
AI landscape specifications and maintenance plans: outdated standards and long-term obligations.
AI in play space and public realm design: injury, accessibility and PI versus public liability.
AI urban greening and green infrastructure scores: wrong surface factors and missed policy targets.
AI cost estimates for landscape schemes: under-pricing and where commercial loss ends and PI begins.
AI images in competitions and marketing: truthful advertising and imitation claims.
Does landscape architects' PI cover AI errors?: Rule 5, TGN 01/13 and silent versus affirmative AI cover.
AI, copyright and landscape design ownership: R4g(c), CDPA s.9(3) and Getty v Stability AI.

For the general insurance questions, see does PI cover AI mistakes? and affirmative versus silent AI cover.

How Apex helps

Apex arranges professional indemnity insurance for landscape architects, LVIA practitioners and landscape design practices. A broker can review how your wording treats AI-assisted work and help you describe that use accurately at renewal.

Sources

Frequently asked

Does the Landscape Institute allow members to use AI?

Yes. The LI has published no AI guidance, policy or Code provision as at October 2026, and nothing in the December 2021 Code of Practice prohibits AI tools. Members remain bound by Rule 6 on competence, Rule 11 on honesty and transparency and R4g on agreements and data, so they must be able to check and stand behind any AI-assisted output.

Do I have to declare AI use in landscape evidence at appeal?

Yes, where the Planning Inspectorate guidance applies. Intentional AI use to draft, summarise, analyse, or generate or alter images or videos must be declared, saying which parts and what checks were made, and confirming responsibility for accuracy. Each expert declares for their own documents, so a landscape witness makes their own declaration.

Is there a minimum PI limit for landscape architects?

No. LI Rule 5 requires appropriate PI covering relevant landscape services, and Registered Practices must carry and evidence PI, but the LI sets no minimum limit or wording. Clients, frameworks and public bodies often set their own requirements in the appointment, which should be checked against your policy.

Will my PI cover a mistake made by an AI tool?

Often it will, because most landscape PI wordings do not mention AI and respond to negligent professional services however they were produced. Cover still depends on the policy terms, any exclusions and whether your AI use was fairly presented to insurers. A broker can check your specific wording.

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.