Our exploration began with general AI tools such as Chat-GPT and Copilot, but we quickly discovered their limitations. While general AI can explain construction terminology, it cannot assess complex development projects with the nuanced understanding of planning risk, construction programming, market absorption rates, and developer track records that our business requires.

As AI adoption increases, we expect more refined agents capable of development-specific analysis – planning risk assessment, construction draw monitoring, market absorption forecasting, and the interplay between development costs and gross realization values.

Our research has revealed several compelling examples of AI implementation across commercial real estate, mainly in north America and mainly applicable to stabilised investment property. CRE Agents is building the first AI Agents for Commercial Real Estate, automating repetitive tasks across 17 functional areas including asset management, acquisitions, and development. LeaseLens offers AI lease abstraction services, while companies like Prophia specialise in AI-powered asset management platforms.

We haven’t found any agents specifically focused on development finance and at this point are investigating building our own internal ‘agent’ that will assist us in pulling data, conducting analysis and providing recommendations in our space. Over time, our expectation is that this will supplemented with more advanced external agents that will amplify the benefits of AI further.

What excites us most is the potential for analysis that would be impossible with current methods. AI could allow our analysts to go deeper into development-specific datasets and modelling techniques that would be prohibitively complex to execute consistently.

Development Project Documentation: We’re researching how AI could review and resolve complex development documentation—construction budgets, construction plans, engineering reports—extracting key parameters and structuring data for comparative analysis.

Construction Cost and Timeline Modelling: Systems we’re exploring could analyze construction cost against historical data, identify potential cost overruns, and forecast program risks incorporating variables like labour costs, material costs, and council approval timelines.

Market Absorption Analysis: AI systems could provide real-time market intelligence, analyzing population growth, demographic trends, competing supply, and buyer behaviour to assess project feasibility based on dynamic market conditions and property cycles rather than current snapshots.

Our most important discovery is that AI’s value lies in enhancement, not replacement. We’re exploring how to elevate our team’s capabilities and focus expertise on the complex qualitative assessments that define successful development investing.

AI could handle data compilation, initial project analysis, and standardised reporting, allowing analysts to focus on assessing developer capabilities, evaluating market strategies, and making nuanced decisions about project viability.

We’re excited about potential applications in construction progress claim administration and project monitoring. AI systems might analyse draw requests against approved budgets, review progress documentation, and flag discrepancies while preserving our focus on relationship management and strategic decision-making.

We recognize that AI represents more than operational efficiency—it could provide competitive advantages through enhanced due diligence speed, superior portfolio monitoring across development types, and improved developer experience through faster, more sophisticated analysis without increased overhead.

Our exploration has revealed both opportunities and risks that inform our approach. Development investing data is more complex and less standardized than investment property information, creating data quality risks that could lead to flawed AI outputs. Each project’s uniqueness presents standardization challenges.

Key risks we’re considering include: Over-reliance on AI recommendations without sufficient human oversight could result in poor investment decisions. Biased algorithms might perpetuate unbalanced analysis if not properly monitored. Data security concerns arise when processing sensitive development project information through AI systems.

Our mitigation strategies focus on: Maintaining human oversight for all critical decisions, implementing robust data validation processes, partnering with AI providers who understand development finance complexities and prioritize data security, and starting with lower-risk applications to build confidence and competence gradually.

We’re developing a clear vision of AI-enhanced development investing. We envision our team spending more time on developer relationships, market strategy, and complex project assessment, supported by AI systems handling routine analysis and providing sophisticated decision support.

Any implementation must enhance rather than replace the deep market knowledge, development expertise, and developer relationship management that define successful development investing.

What excites us most is the opportunity to approach AI strategically. We’re not rushing to adopt AI but carefully considering how it could address specific challenges in development investing.

At Ark, we’re energised by the discovery process itself. We’re actively exploring how to shape our future through intelligent AI integration designed for the unique challenges of land development and construction financing. The questions we’re asking will determine how effectively we harness AI’s potential to transform how development finance and investment decisions are made, risks assessed, and projects monitored.

The exploration continues, and the possibilities excite us.

If you’re an investor, borrower or strategic partner who values forward-thinking leadership, now is the right time to connect.

Let’s talk about where you want to go and how we can help you get there.

Want more articles like this? Follow Peri Macdonald on LinkedIn.