By Bjorn Fehrm and Henry Tam

February 6, 2026, ©. Leeham News: We have completed a detailed, step-by-step analysis of the certification requirements a Part 25 Air Transport airliner in the 200-seat segment must meet.

In our series, we have seen work that could benefit from an AI agent, and other work where we conclude it will be difficult.

We begin this week by outlining areas where we expect AI to reduce the number of work hours required to complete a task. We will attribute these AI-driven work-hour reductions to the appropriate areas of the aircraft Program Plan in Figure 1.

Figure 1. A generic new Part 25 airliner development plan. Source: Leeham Co. Click to see better.

 Work that AI can speed up

Over the past few years, AI technologies advanced significantly.  It can now generate documents, images, and videos.  Agentic AI can even handle (to some extent) multi-step workflows. What are some areas that AI could help to save work hours?

Documents

One area where AI could help is template creation.  Templates can affect thousands of documents to be released during a development program.  A good template could reduce review time by presenting information consistently.

AI could generate these templates relatively quickly.  Users may still need to fine-tune the output to ensure that the templates are suitable for their applications.

In addition to templates, AI can also help with definitions and acronyms lists.  These sections are important and often time-consuming for engineers to compile.  AI can probably do this faster than its human counterpart.

Furthermore, AI could perform basic quality control of these documents.  Companies often have standards for document format, the use of certain terminology, how graphics are labelled, etc.  If properly trained, AI could perform some of these tasks, freeing up staff for more value-added work.

Requirements Development

AI can also help with requirements development.  Current technology can help engineers quickly identify applicable airworthiness rules for components or subsystems, in our case, in the relevant Part 25 sections.   It can also provide justification or explanation for why these rules are applicable.  Yet, engineers still need to review and confirm that the information is correct.  The person who signs off on the document remains responsible for it.

Structural Design

An airliner in the class we are analyzing contains about four million parts. About half of these are standard parts, such as rivets, fasteners, and fittings.  There is also a large number of small parts that are bespoke, such as brackets, clips, etc.

AI could help design simpler parts, conduct relevant analyses, and generate reports for engineers to review. AI-enabled mechanical 3D CAD has advanced to the point where the work-hours required to create these brackets should be substantially reduced.

Support Meetings

With current technologies, AI can be used to take notes and summarize meetings.  It has the potential to generate meeting minutes, reducing the workload of engineers and project managers.  Keep in mind that the team still needs to ensure all key points and actions are captured.  Additional word crafting may be required to handle sensitive topics.

Drafting Documents with an Abundance of Publicly Available Examples

AI could also help draft documents with lots of publicly available training data.  For example, AI could help managers draft job descriptions, so they just need to fine-tune the text instead of starting from scratch.  The tool would save some hours of work and facilitate the recruiting process.

Where AI helps

These are the areas where, in discussions with experienced airliner developers who contributed to the articles, it was clear that AI assistance would help. Next week, we will review areas where it’s more difficult to see a fit for AI assistance.