Insights
AI in aviation compliance, explained by an engineer
Guides on what AI can and cannot do in certification, airworthiness and operations documentation. Each article is written by Oguz Hicdurmaz, cites the regulator source it relies on, and carries the date it was last checked.
· 8 min read
How to choose an AI tool for aviation compliance
How to choose aviation compliance software with AI: the four kinds of tool that answer the same search, ten checks that separate a cited EASA or FAA answer from a plausible one, a 30-minute test to run on any tool with your own documents, and how Wingman360 Teammate answers each check.
Read the article· 9 min read
Can AI be trusted for aviation compliance documentation?
An airworthiness engineer answers the question every aviation organisation is asking: what AI can safely do in compliance documentation, where it fails, what EASA and the FAA say, and the four conditions under which its output can be relied on.
Read the article· 11 min read
Certification-basis compliance matrix: EASA Part 21 and FAA
A step-by-step guide from an airworthiness engineer: what EASA Part 21 and FAA Order 8110.4C expect a compliance checklist to contain, the columns that keep a matrix usable through the programme, the eight steps to build one, and which steps AI can draft and which an engineer must decide.
Read the article· 12 min read
FAA aircraft compliance checklist under 14 CFR Part 21
How to build the compliance checklist FAA Order 8110.4C asks for in the certification plan: the certification basis under 14 CFR 21.17 with amendment levels, special conditions, ELOS findings and exemptions, the means of compliance the order names, the compliance report behind each row, the statement of compliance under 21.20, and which steps an AI teammate drafts and which the applicant decides. With the scope rules for changes and STCs under 21.101.
Read the article· 12 min read
STANAG 4671 compliance matrix for a military UAS: AEP-4671
How a compliance matrix against the Unmanned Aircraft Systems Airworthiness Requirements (USAR, AEP-4671) is built inside an EMAR 21 certification programme: what STANAG 4671 Edition 3 is, how the national military airworthiness authority establishes the certification basis under EMAR 21.B.80 by tailoring EMACC, the columns a military matrix needs, seven steps, and which of them an AI teammate can draft on the organisation's own hardware. Quotes no USAR text.
Read the article· 13 min read
EASA SC Light-UAS and SC-VTOL: means of compliance with AI
How the means-of-compliance table and the design verification evidence are built for an unmanned aircraft under SC Light-UAS or an eVTOL under SC-VTOL: what the two objective-based special conditions cover, how the SORA and SAIL decide between a statement of compliance, an EASA design verification report and a type certificate, the five items a proposed means of compliance must state, what the DVR application asks for, seven steps, and what an AI teammate drafts while the margins, criteria and classifications stay with the engineer.
Read the article· 13 min read
SORA 2.5 with AI: the ten steps, the UAS operations manual
The specific operations risk assessment as EASA adopted it in AMC1 to Article 11 of Regulation (EU) 2019/947 (JARUS SORA 2.5, ED Decision 2025/018/R): what Article 11 and points UAS.SPEC.030 and UAS.SPEC.050 require, what each of the ten steps decides and produces, the seventeen operational safety objectives, the Annex A structure of the UAS operations manual and the compliance matrix, the comprehensive safety portfolio, seven steps from concept of operations to application, and what an AI teammate drafts while the claims, the robustness and the application stay with the operator.
Read the article· 11 min read
AI assistant for EASA Part 145 and CAMO organisations
What an AI assistant can do inside a maintenance organisation or a CAMO: answer staff questions from the MOE or CAME with the paragraph cited, draft amendments and audit responses for review, and where the line sits under points 145.A.70, 145.A.45, 145.A.55, 145.A.200 and CAMO.A.300. With six steps to introduce one.
Read the article· 12 min read
DOA handbook, MOE and CAME with AI: what EASA requires
What the design organisation handbook (Part 21, point 21.A.243), the maintenance organisation exposition (Part 145, point 145.A.70) and the continuing airworthiness management exposition (Part-CAMO, point CAMO.A.300) must contain, the AMC layouts, who signs, which changes need prior approval under 21.A.247, 145.A.85 and CAMO.A.130 and the time frames, seven steps to draft or rebuild one, and what an AI teammate drafts while the statement and the nominations stay with the accountable manager.
Read the article· 10 min read
On-premise and air-gapped AI for aviation and defence
What Part-IS (Regulations (EU) 2023/203 and 2022/1645) asks an aviation organisation to assess before it adopts an AI assistant, the four deployment models and the interfaces each one creates, what air-gapped operation really requires, and seven questions to put to any vendor.
Read the article· 11 min read
Functional hazard assessment with AI: ARP4761A, AC 25.1309
What ED-135 and SAE ARP4761A expect an aircraft or system FHA to produce, the failure condition classification scale and its probability objectives under AC 25.1309-1B, the six places FHAs go wrong, and which of the six AFHA activities AI can support without taking the classification away from the engineer. With a note on unmanned aircraft under ED-279.
Read the article· 9 min read
Wingman360 vs ChatGPT, Claude and Gemini for aviation work
A criterion-by-criterion comparison of the general assistants and an aviation knowledge assistant: where answers come from, citations, revision control, who can change the corpus, training on your data per OpenAI, Anthropic and Google's own statements, deployment, security attestations, document workflows and price. With a clear account of what each is better at and how organisations use both.
Read the article