Where AI is already executing work in aerospace

Artificial Intelligence (AI) is often discussed in enterprise terms as a support layer that helps teams move faster and make better decisions. But in industries such as aerospace, the shift is already going further. At GE Aerospace, AI is not just generating insights. It is beginning to shape design choices, sharpen inspection, predict maintenance, and improve operational readiness across the lifecycle of an engine. In this Dataquest Q&A, Jayanth Sekar, Sr. Director, Data Science & AI – Customer Experience, GE Aerospace, explains where AI is already executing work, and why human control, transparency, and trust remain essential in critical environments.

Where is GE Aerospace already seeing AI move beyond support and into execution?

At GE Aerospace, AI has been in use for more than a decade, so this is not a new experiment for us. What is changing now is the depth of its role. We are seeing AI move from insight and support into execution across the full engine lifecycle, from design and production to maintenance and service readiness.

One clear example is engine design. Our researchers have demonstrated generative AI systems that can produce hundreds of design iterations in seconds. That has the potential to shorten development cycles for future engine technologies in a meaningful way. It marks a clear shift from AI simply assisting engineers to AI actively participating in the design process.

We are also seeing AI work in execution mode when it comes to fleet readiness. We monitor around 50,000 commercial engines globally, 24/7, using AI to identify predictive maintenance issues. This has led to a 45% increase in issue detection, 60% faster lead times in identifying maintenance needs, and a 50% reduction in false alerts.

Another strong example comes from inspection. The Blade Inspection Tool developed by our engineers at the GE Aerospace Bengaluru Technology Centre has cut inspection times by half. We are also using AI to forecast work requirements for engine maintenance, so that the required parts are available before an engine even reaches a Maintenance, Repair, and Overhaul facility. That helps reduce service delays and return engines to aircraft sooner.

There is also a workforce productivity layer. Our internal generative AI platform, AI Wingmate, is now used daily by around 11,000 employees. It helps streamline routine work and supports learning, all within a protected network.

What has to be solved before enterprises can truly trust AI as an execution layer?

The most important issue is unmanaged autonomy. In critical industries such as aerospace, the goal cannot be unchecked automation. It has to be augmentation. That is why our approach to scaling AI is fundamentally human-centric. We build co-pilots, not replacements, so that expert humans with deep domain knowledge remain central to innovation and in control of critical decisions.

For enterprises to trust AI in an execution role, there has to be a strong foundation of responsible AI principles. For us, that rests on three pillars: trust, transparency, and human centricity.

The first is trust. We take a highly methodical approach to selecting the right data for model training. Over time, we have invested in bringing together more than 30 years of complex engineering and service data. That gives our models a depth of experience that would be difficult to replicate in any other way.

The second is transparency. We build trust in our AI systems through a rigorous validation approach centred on explainable AI. Our models are designed to show their work and provide the reasoning and data behind their recommendations. That level of transparency matters both for our teams and from a regulatory standpoint.

The third is human centricity. AI can enable faster decisions, but the final call still has to rest with humans. Our Engineering Assistant is a good example. It acts as a partner to engineers by drawing on decades of service data, helping them solve problems faster, and generating hundreds of potential design options or solutions. But the final decision remains with the engineer.

Ultimately, this balance is grounded in a culture of safety and quality. We apply the same engineering discipline to AI that we apply to our engines. That is how accountability and safety stay intact, even as AI takes on a more active role in execution.