Azzera builds AI flight data checker for aviation compliance

Azzera is developing an AI verification layer for its CELESTE platform that cross-checks operator flight data against external sources to catch discrepancies before audits do. The Montréal company says the tool is built with strict data-confidentiality controls as emissions and SAF reporting rules tighten across Europe and beyond.

Why it matters: - Aviation operators face rising compliance pressure across the EU ETS, UK ETS, Swiss ETS, CORSIA and ReFuelEU. - ReFuelEU’s sustainable aviation fuel mandate rises from 2% in 2025 to 6% by 2030. - CORSIA becomes mandatory for most member states in 2027. - A faster check on flight data can reduce audit delays, manual reconciliation and the risk of late-stage findings.

What happened: - Azzera is developing an AI-powered verification layer for its CELESTE platform. - The tool cross-checks operator flight data against authoritative external sources, including EuroControl, NAV Canada and CRCO. - The system is designed to flag discrepancies while operators are preparing reports. - The rollout is starting with testing among existing CELESTE clients. - Azzera plans to expand coverage over time before a full rollout to more than fifty aircraft operators in Europe and North America.

The details: - The verification layer identifies flights that appear in one system but not another. - The tool also flags mismatches in flight-level data between an operator’s records and independent sources. - The AI layer generates plain-language explanations for why a discrepancy occurred. - Azzera says the tool can turn hours of investigation into an immediate answer for the team reviewing a report. - The verification layer runs inside a contained process built only to compare operator data against approved sources. - Client data stays inside each operator’s own CELESTE account. - Client data is not pooled across customers. - Client data is not used to fine-tune, train or benchmark the model. - The same data-handling standard applies across Azzera’s platform, including the SAF POD beta program. - Azzera is receiving advisory services and funding from the National Research Council of Canada Industrial Research Assistance Program to support the research and development project. - The project focuses on flight data verification and compliance reporting. - The company also cites prior funding from Sustainable Development Technology Canada and Innovative Solutions Canada. - Azzera says the project builds on its ongoing INSAT-funded SAF POD initiative. - More information is available in the company's announcement.

Between the lines: - The move shows how aviation compliance software is shifting from manual review toward automated pre-checks. - The confidentiality approach is the centerpiece of the product pitch because compliance data can carry financial and legal consequences. - The emphasis on a contained process suggests Azzera is trying to answer one of the biggest objections to AI in regulated workflows: whether sensitive client data could be reused to train models. - The company is framing safe AI as a competitive advantage in aviation decarbonization, not just a technical feature.

What's next: - Azzera will expand the verification layer’s data-source coverage beyond its first set of external inputs. - The company is moving toward a broader release across its operator base after the current testing phase. - The tool will likely become more relevant as emissions reporting and SAF obligations tighten further in Europe.

The bottom line: - Azzera is trying to make flight-data compliance faster and safer by pairing AI verification with strict data controls.