The math for decarbonizing a real estate portfolio is simple, if you ignore the part where you need to know what you own. For a property manager staring at a spreadsheet of aging buildings, the missing data points,erroneous energy ratings, undocumented insulation, hidden roof structures,can stall a retrofit plan for years. ADAPT AI, a Paris-based startup, is betting that machine learning can fill in those blanks, generating rehabilitation plans and solar potential assessments without dispatching a single engineer with a clipboard [ADAPT AI, retrieved 2024].
The AI Wedge for Stalled Retrofits
The company’s core proposition is a data-cleaner and a planner rolled into one. Its software ingests whatever portfolio information a real estate company has, then uses algorithms to predict missing or erroneous data [ADAPT AI, retrieved 2024]. The goal is to move from a state of uncertainty to a prioritized, actionable decarbonization roadmap. This is a classic climate tech wedge: the unit economics of sending surveyors to hundreds of sites can kill a project’s viability before it starts. By offering a software layer that works with existing, imperfect data, ADAPT AI aims to lower the initial cost and time barrier to creating a portfolio-wide strategy. The tool also promises to scan for rooftop solar potential at scale, a feature highlighted in a speaker profile for an AI for Good event [AI for Good - ITU, Unknown].
An Early-Stage Bet on a Growing Niche
Founded in 2023, ADAPT AI is navigating a European market where regulatory pressure on building emissions is intensifying. The company appears to be in a very early, perhaps bootstrapped, phase. There is no public record of funding rounds, named founders, or customer deployments. What exists is a clear product description and a focused mission: using AI to make profitable investments in the ecological transition [ADAPT AI, retrieved 2024]. The company’s public footprint includes a website, an iOS app, and an open role for an undefined position based in Paris or Geneva, suggesting active development [Stage.fr, Unknown]. The competitive landscape includes more established players like Deepki, which offers a broader sustainability data platform for real estate.
The Credibility Hurdle
For any software promising to make high-stakes capital investment recommendations, the primary risk is accuracy. A property owner will not greenlight a million-euro facade renovation based on a probabilistic model alone. ADAPT AI’s answer, implied by its marketing, is that its predictions are sufficiently reliable to guide the initial audit and planning phase, saving time and cost before more detailed, site-specific studies are commissioned [ADAPT AI, retrieved 2024]. The company’s success will hinge on proving that its algorithms can consistently deliver plans that are both financially sound and compliant with evolving regulations like the EU’s Energy Performance of Buildings Directive.
Running a back-of-envelope calculation illustrates the stakes. If a traditional audit for a 100-building portfolio costs €1,000 per site, the upfront spend is €100,000. If ADAPT AI’s software can cut that initial assessment cost by 80%, the savings of €80,000 could be redirected into the first phase of actual retrofit work. That’s the kind of math that gets a CFO’s attention. To win, ADAPT AI must prove its models are more than a digital guess,they must become the trusted first step that replaces the old, slow, and expensive manual survey. Its real competition isn’t just other software; it’s the inertia of the status quo and the consultant with a clipboard.
Sources
- [ADAPT AI, retrieved 2024] Adapt AI | Transition énergétique | Plan de décarbonation | https://www.adapt-ai.fr/en
- [AI for Good - ITU, Unknown] ADAPT AI - AI for Good - ITU | https://aiforgood.itu.int/speaker/adapt-ai/
- [Stage.fr, Unknown] Postes chez Adapt ai | Stage.fr | https://www.stage.fr/company/136681/adapt-ai/