From Design Toy to Operations Engine
In many design firms, AI is moving from the studio floor to the back office. Landscape architecture practices that once experimented with AI mainly for visual concept generation are now deploying it to streamline text-heavy, operational work. At TBG Partners, president Bill Odle describes how the firm “jumped in with both feet” and now treats AI as a way to augment teams and free designers to focus on relationships and creative problem-solving. Others, like SiteWorks, use tools such as ChatGPT to draft responses to requests for proposals (RFPs), generating a rough structure that staff then heavily edit. Researchers observing this shift report that firms increasingly rely on workflow automation AI for briefs, proposals and internal documentation. The key pattern: AI back office tools are quietly becoming part of everyday practice, helping creative studios automate the boring but necessary work that sits around the real design.
How Landscape Firms Use AI Back Office Tools in Practice
Interviews with landscape architects show AI admin apps being used for specific, often unglamorous tasks. SiteWorks feeds RFP documents into ChatGPT, then refines the 20 percent of usable content it produces to save time on structuring and drafting responses. At Henning Larsen, teams upload large client briefs into tools like NotebookLM and Perplexity to power document ingestion and search. Instead of trawling through hundreds of pages to find a specification, staff ask natural-language questions such as “How wide should the bike lane be?” and get the answer, with source references, from within their own document set. For multilingual or text-averse colleagues, the same data can be transformed into summaries or even podcast-style briefings. This kind of creative studio automation targets project management, client communication and documentation—areas where consistent information and speed matter more than originality, and where human review remains a mandatory final step.
Oberlin’s AI Micro Grants: Administrative and Research Support
Oberlin College’s AI Micro Grant Program offers a parallel case in academia: small, time-limited funding for staff and faculty to explore AI in research and administration. Projects span disciplines from Dance to Environmental Studies but share a focus on workflow automation AI rather than flashy content generation. Assistant Professor Al Evangelista uses AI tools to process large text databases of fact-checked claims from South and Southeast Asia, examining how AI reshapes the spread of misleading content across languages and regions. Environmental studies professor John Petersen is testing AI to gather and organise local civic event data into a unified community calendar, aiming to improve information sharing and support more sustainable communities. These AI back office tools are being evaluated as service infrastructure: ways to handle data processing, scheduling and information management so that academic staff can spend more time on teaching, community work and deeper research.
Benefits, Risks and the Need for Deliberate Adoption
Across both design firms and universities, the benefits of AI in design firms’ back offices cluster around time savings, reduced repetitive tasks and better tracking of information. Designers report that automating “nug work” on RFPs preserves their creative energy for higher-value thinking, while document search tools help teams move faster without losing important details. Yet researchers warn that many organisations fail to measure impact, treating AI like a vague productivity booster rather than a tool that should be evaluated like any new hire. Concerns include accuracy in AI-generated summaries, over-reliance on generic outputs and the environmental costs of large-scale AI usage. Effective use of AI admin apps requires human oversight, clear processes and awareness of data sources. Without training and governance, workflow automation AI risks becoming a black box that staff neither fully trust nor understand, especially in fields where precision and ethics are central.
Practical Paths for Malaysian Studios and Agencies
For Malaysian creative studios and knowledge-focused agencies, these case studies offer a pragmatic roadmap. Rather than aiming for full automation, start with contained pilots: AI meeting note-takers that summarise client calls, tools that extract key requirements from RFPs, or invoice processing systems that classify expenses for finance teams. Simple agents can help manage project documentation, pulling the latest drawings, briefs and emails into a searchable workspace. Inspired by Oberlin’s micro grants, organisations can set up small internal funds or time allocations for teams to test AI back office tools, with clear objectives and success metrics. Crucially, involve staff in choosing and refining AI admin apps, and pair every deployment with training on prompts, verification and data handling. By treating AI as a quiet assistant for admin and operations, Malaysian studios can reclaim time for design, research and client relationships—the work humans do best.
