Technology Readiness Levels: Are TRLs Still Fit for Purpose?

Rethinking Technology Maturity in the Age of AI, Autonomy and Rapid Innovation

Technology Readiness Levels (TRLs), originally developed by NASA, remain widely used across aerospace, defence, engineering and publicly funded innovation programmes. The nine-level framework provides a common language for assessing technological maturity, from basic scientific principles (TRL 1) to proven operational performance (TRL 9).

But does a framework developed around relatively predictable engineering development still reflect how emerging technologies evolve?

Traditional TRLs work particularly well for hardware, materials, propulsion systems and physical prototypes. However, artificial intelligence, autonomous systems and continuously updated software introduce different challenges.

An autonomous vehicle may demonstrate reliable operation in one environment but fail in another. An AI model may achieve excellent laboratory performance yet deteriorate when confronted with unfamiliar conditions. Software updates can fundamentally change system behaviour without changing the physical platform.

TRLs also measure technology maturity, not necessarily manufacturing readiness, system integration, cybersecurity, regulatory acceptance or commercial viability.

Complementary frameworks already exist, including Manufacturing Readiness Levels and Integration Readiness Levels. The opportunity may therefore lie in combining assessments rather than replacing TRLs entirely.

A more adaptive approach could evaluate operational resilience, autonomy assurance, software evolution and deployment readiness alongside conventional technical maturity.

TRLs remain useful, but a single maturity number increasingly risks concealing the very uncertainties that determine whether advanced technologies succeed.