The era of greenwashing may be coming to an end, at least in the construction sector, where manufacturers are embracing a low-cost tool to support data-driven transparency about their environmental impacts. Lodestellar, an Estonian startup, uses artificial intelligence to automate the quality checking of Environmental Product Declarations (EPDs), helping companies avoid costly delays, compliance risks, and verification expenses.
EPDs are standardized documents that disclose the environmental footprint of construction products, from raw material extraction to manufacturing and disposal. They rely on Life Cycle Assessment (LCA) methodology and independent verification to provide credible data. As global regulations tighten—such as the European Commission's new EU-wide framework for calculating building life cycle impacts—specifiers increasingly select suppliers based on EPD data. Without an EPD, manufacturers face assumed carbon values that overestimate their product's impact, putting them at a severe disadvantage in bids for large infrastructure and commercial projects.
Around 50 EPDs are published daily in the construction sector, an exponential rise that is overwhelming independent verifiers. Creating a single EPD can cost over €10,000 and take more than six months, often involving multiple revision rounds during verification. If the declaration fails to meet standards, the entire investment may be lost. Lodestellar addresses this bottleneck by acting as a "linter" for EPDs—similar to automated code-checking tools used by software developers—providing line-by-line guidance on draft declarations before they are submitted for formal verification.
From Hackathon to Industry Standard
Lodestellar was born in 2022 at a digital construction hackathon organized by the Estonian Economic Ministry, which aimed to spark game-changing digital tools for the global construction sector. A panel of industry experts awarded first place to the concept, which was then developed by a team of LCA experts led by CEO Anni Oviir, herself a certified EPD verifier and LCA lecturer who co-authored Estonia's national carbon footprinting method. The technical lead is CTO Tanel Teinemaa, an experienced software developer behind some of Estonia's leading tech companies. The tool was refined through extensive validation, including use at LCA Support, and rolled out widely in March 2024.
"We've examined years of EPD verifier feedback," Oviir said. "The quality issues that cause delays tend to follow a pattern: missing explanations, ambiguous assumptions, inconsistent wording, omitted mandatory disclosures, and poor traceability. Our tool automates the detection of these problems, allowing EPD professionals to focus on higher-level considerations."
Lodestellar reads an EPD—whether in draft form or already published—automatically detects which standards apply (EN 15804+A2 or ISO 21930), and returns a detailed report with clear guidance. It runs hundreds of different checks, sometimes being "a bit too nitpicky" but never missing an issue it has been instructed to find. The team has tested hundreds of EPDs and found that not a single one—even among those already published—was free of issues that could have been spotted earlier.
Industry Validation and Future Plans
Industry experts have praised Lodestellar's role. Dr. Roger Singleton, CEO of Riskoa and operator of the EmVide AI LCA platform, said: "As AI accelerates the production of product-level climate data, independent quality review becomes essential. Lodestellar adds an independent layer, so LCA software providers aren't marking their own homework." Professor Callum Hill, an independent LCA expert and EPD verifier, added that the tool "acts like an extra pair of eyes" and helps catch issues that are easy to miss in complex models.
Lodestellar is now used by some of the biggest names in the building product industry. The team is developing new features, including the ability to analyze EPD background reports, and internal benchmarking methods that will be shared publicly to transparently track performance improvements. Pricing remains low—starting at €7 per check—to keep the tool accessible and support quality at scale.
The broader context is that construction accounts for approximately 40% of global greenhouse gas emissions, making it a priority sector for carbon accountability. Policymakers worldwide want to cement EPDs as a standard practice, incentivize low-carbon production, and use the data in initiatives like the EU's Digital Product Passports. As AI advances, buyers—specifiers in construction—can more easily weigh complex factors like carbon cost, fire-proofing, acoustics, and financial costs across all products on the market. This shifts the advantage to companies with credible, verified data rather than vague eco-slogans.
Oviir emphasized: "The companies at the forefront of extracting value from AI are also the companies that never use it as a buzzword. In both AI and sustainability, results must speak for themselves." Lodestellar aims to raise EPD quality beyond compliance, ensuring that manufacturers can compete on a level playing field with trustworthy environmental disclosures. With the tool, developers can fix basic issues earlier, reduce back-and-forth with verifiers, and ultimately deliver results faster—helping them win those multi-million euro tenders.