AI in aerospace and defense: two studies, a 9x gap for the same year

Two reputable research firms have published, months apart, two estimates of the global “AI in aerospace” market that diverge by a factor of 9 for nearly identical reference years. SkyQuest Technology Consulting (2026 report, already cited in our 23/08 analysis) prices the market at $3.28B for 2025. ResearchAndMarkets, in a report published in January 2026 covering “AI in aerospace and defense,” values it at $29.27B for the same year — trending toward $33.31B in 2026 and $55.3B by 2030 (13.5% CAGR).

The gap isn't a forecast disagreement in the usual sense: the two firms aren't disputing the trajectory of the same market, they simply aren't measuring the same object. SkyQuest's scope appears limited to AI software applied to civil aerospace. ResearchAndMarkets aggregates aerospace AND defense, and likely includes embedded systems, avionics, and hardware platforms far costlier than software alone — a framing consistent with the deals it cites (Safran's $254M acquisition of Preligens, Safran Electronics and Defense's Advanced Cognitive Engine), both defense-oriented.

We documented a comparable forecast-trajectory divergence on 14/08, in upstream oil and gas AI, between two research firms — but that was a disagreement over growth rate from a comparable base. Here, it's the starting measurement itself — today's market size — that differs by a factor of 9: a definitional problem, not statistical uncertainty.

The practical takeaway for any executive citing these figures in an AI business case: the headline TAM can swing 9x depending on which firm you pick, with forecast quality having nothing to do with it. The question to ask before citing a number isn't “is the forecast good?” but “what exactly does this number count?”