What thirty years of biomimetic disappointment can teach us about reading the field’s genuine crossings

There is a particular kind of technology that has been eighteen months from changing everything for the better part of forty years. Biomimetics — the discipline of borrowing nature’s solved engineering problems — is the standing example. Sharks for drag, geckos for adhesion, insects for flight, the mantis shrimp for optics: the field has produced an unbroken stream of arresting demonstrations and a remarkably thin record of fielded capability. For anyone whose job is to distinguish what is coming from what is merely recurring, biomimetics is a useful discipline precisely because it is a graveyard of premature announcements. It teaches the reader to ask the only question that matters: has this particular borrowing actually crossed from the laboratory into something that works under load, in the field, at cost — or is it once again beautiful in a wind tunnel and nowhere else?

This Signal maps where the field actually stands. It does three things. It states plainly why biomimetics has underdelivered relative to its promise. It identifies the small number of areas that have genuinely crossed, or are close to crossing, into capability — and it is honest about how few they are. And it flags the three that a defence-technology reader should watch, with an explicit readiness assessment for each, because the entire value of following this field lies in telling the crossing from the demonstration.

Why the borrowing so often fails

The romance of biomimetics is that evolution has already done the engineering. Four billion years of iteration have produced surfaces, structures, and control systems that human designers can, in principle, simply copy. The reality is that biology and engineering optimise against different constraints, and the gap between them is where most biomimetic programmes quietly die.

Consider the gecko. The mechanism by which a gecko’s foot adheres to a wall — van der Waals forces across millions of microscopic setae — has been understood for two decades. Reproducing the adhesion in the laboratory is straightforward. Reproducing the gecko’s other trick, the one that matters, is not: the animal’s foot stays clean, shedding the dust and grime that would otherwise clog the setae and destroy the adhesion within minutes. The self-cleaning is the hard part, and it is the part that determines whether a gecko-inspired adhesive is a laboratory curiosity or a usable material. Most biomimetic programmes reproduce the easy mechanism, announce a breakthrough, and founder on the constraint that biology solved and the copy did not.

The pattern generalises. Biological structures are grown, not manufactured, and the growth process embeds tolerances, gradients, and self-repair that are extraordinarily difficult to reproduce with human fabrication. A biological surface that works is the product of a whole organism maintaining it; a copied surface has to survive on its own, without the metabolism that kept the original clean, intact, and tuned. This is why the field’s failures cluster not at the moment of demonstration — demonstrations are easy — but at the transition to durability, manufacturability, and cost. The wind tunnel is not where biomimetics dies. The maintenance schedule is.

The strategic lesson is that military interest in a biomimetic concept tells you almost nothing about whether it will field. Defence research organisations have funded gecko adhesives, insect-inspired micro-air-vehicles, and cephalopod-inspired camouflage for years, and the funding is not evidence of imminent capability. It is evidence of enduring interest in a hard problem. Reading the funding as a readiness signal is the single most common error in this field, and it is the error this Signal is written to prevent.

Where the field has actually crossed

Against that sceptical backdrop, one biomimetic technology has genuinely crossed into operational use, and it is worth being precise about it because it is the exception that calibrates the rest. Shark-skin-inspired riblet films — microscopic grooves, tens of micrometres high, that replicate the drag-reducing dermal denticles of shark skin — are now flying on commercial aircraft in revenue service. The Lufthansa-BASF product, AeroSHARK, has been applied across an entire long-haul fleet; a Boeing 777 operator equipped all of its aircraft of that type over a twenty-month programme, and other carriers have followed. A separate riblet film, developed in Australia, was applied to a supersonic demonstrator aircraft in 2024. These are not wind-tunnel results. They are films on airframes carrying passengers and freight, surviving the maintenance cycle, delivering a measured single-digit-percentage reduction in drag.

Epistemic status: this crossing is Confirmed. The fleet applications are documented from primary manufacturer and operator sources; the drag-reduction figures (of the order of one per cent at the fleet level, higher on optimally treated surfaces) are reported consistently across independent operators. This is a genuine TRL 8–9 technology in the civil-aviation configuration.

The reason riblets crossed when gecko adhesives did not is instructive, and it is the key to reading the rest of the field. Riblets are passive. They have no moving parts, require no self-cleaning to function, and impose no maintenance burden beyond the film’s own durability. The borrowing succeeded because the biological function being copied — a static surface geometry that reduces turbulent skin friction — is one of the rare cases where the easy-to-reproduce mechanism is also the whole mechanism. There is no hidden constraint, no metabolic upkeep, no self-repair that the copy has to substitute for. When a biomimetic borrowing has that character — passive, static, self-sufficient — it can cross. When it depends on the organism actively maintaining the function, it usually cannot.

The defence relevance of the riblet case is direct but bounded. The same drag reduction that saves fuel on a civil airliner applies to military transport, maritime patrol, and tanker aircraft, and to ship hulls, where the antifouling property of the shark-skin surface adds a second benefit. This is a real, near-term efficiency gain available now, and it is the kind of unglamorous crossing that actually changes operating costs at fleet scale. It will not win a war. It is, however, the only biomimetic surface technology of which it can honestly be said that the crossing has happened.

Three to watch — with the readiness stated

Beyond the riblet case, three biomimetic areas are close enough to capability, or moving fast enough, to warrant a defence reader’s attention. In keeping with this publication’s standard, each is stated with an explicit technology-readiness assessment, because the assessment is the information. A reader who takes away only the three names has taken away nothing; the value is entirely in knowing how far along each actually is.

The first is bio-inspired legged locomotion. Quadruped robots that borrow the gaits and compliant limb mechanics of animals have moved from research novelty to a real commercial market, with documented interest in hazardous-terrain reconnaissance, inspection, and logistics — precisely the missions where wheeled and tracked platforms struggle. The core mobility is mature. But the specifically biomimetic refinements that the research literature is most excited about — compliant leaping mechanisms borrowed from jumping insects, energy-storing tendon analogues, neural-oscillator gait controllers — remain at the prototype stage, demonstrated on testbeds rather than fielded.

Epistemic status: mixed, and the distinction matters. General quadruped mobility is at TRL 7–8 and commercially available. The biomimetic enhancements that would make these platforms markedly more capable — explosive jumping, tendon-like energy recovery, animal-like terrain adaptation — are at TRL 4–5, demonstrated in the laboratory and on analog testbeds but not in operational configurations. Anyone told that ‘biomimetic robots’ are field-ready is being sold the mature mobility and the immature enhancements as a single package. They are not the same technology.

The second is biomimetic drag and antifouling surfaces beyond the aircraft case. The riblet crossing in civil aviation is real; the question worth watching is how far it extends to naval hulls and to high-temperature applications such as engine components, where the same geometry must survive far harsher conditions. Marine riblet surfaces combining drag reduction with resistance to biological fouling have shown strong results, including substantial reductions in both drag and marine organism adhesion, but largely in test configurations rather than on operational vessels at scale.

Epistemic status: Probable and advancing. Aviation riblets are TRL 8–9, as noted. Marine-hull and high-temperature riblet applications sit at roughly TRL 5–6 — demonstrated in relevant environments, not yet fielded at fleet scale. The trajectory is favourable because the underlying mechanism is the same passive, static geometry that succeeded in aviation; the challenge is durability in a more punishing medium, which is an engineering problem rather than a scientific one.

The third is biomimetic sensing. Sensors that borrow biological architectures — compound-eye wide-field vision, the polarisation sensitivity of the mantis shrimp, insect-inspired collision-avoidance circuits, distributed tactile arrays modelled on skin — are the subject of an accelerating research literature, and the appeal for autonomous systems is obvious: biological sensors achieve extraordinary performance at extraordinarily low power and weight. This is the area where the gap between the demonstrated and the deployable is widest, and where the temptation to over-read is greatest.

Epistemic status: Possible, and mostly early. The most promising biomimetic sensing concepts sit at TRL 3–4 — validated as components in the laboratory, not integrated into fielded systems. This is exactly the readiness band where this field has historically generated its most confident and least reliable announcements. The concepts are real and worth tracking. The capability is not close. Both statements are true at once, and saying only the first is how a serious reader is misled.

How to read the field

The through-line of these assessments is a single discipline. Biomimetics rewards the reader who asks not ‘is this remarkable?’ — it is almost always remarkable — but ‘has the borrowing crossed the specific constraint that biology solved and the copy must now solve without the organism?’ Where the borrowed function is passive and self-sufficient, as with riblets, the crossing can happen and sometimes has. Where the function depends on active maintenance, self-cleaning, or self-repair, as with most adhesives, sensors, and actuated mechanisms, the demonstration is easy and the crossing is hard, and decades of funding have not closed the gap.

For a defence-technology reader, this yields a practical rule. Treat every biomimetic claim as a demonstration until the readiness level is established, and treat the readiness level, not the funding or the elegance of the mechanism, as the only reliable signal. The field will continue to produce breakthroughs that are eighteen months from changing everything. A small number of them — passive, static, self-sufficient — will actually cross. The riblet on the airframe is what a crossing looks like. The gecko adhesive that has been imminent for twenty years is what the other kind looks like. The whole skill is telling them apart before the announcement, not after.

References

Fielded and commercial sources

Lufthansa Group Innovation Runway. “AeroSHARK.” Manufacturer documentation (BASF / Lufthansa Technik). Reports full AeroSHARK equipment of the SWISS Boeing 777-300ER long-haul fleet over a twenty-month programme.

Boom Supersonic. “From Sea to Sky: the Aerodynamics of Sharkskin.” FlyBy Blog, 2025. Documents application of MicroTau riblet material to the XB-1 demonstrator in autumn 2024.

AVweb. “Taking a Tip from Sharks: Advanced Aircraft Coatings Take a Bite Out of Drag,” 2025. EVA Air and All Nippon Airways adoption of AeroSHARK.

AFCEA International, SIGNAL Media. “Animal-Inspired Robots May Soon Take Over World Markets,” December 2025. Quadruped and biomimetic robot market figures and mission set.

Peer-reviewed and technical literature

Systematic mapping review. “Biomimetics as a Functional Engineering Framework for Mechanical Systems: A PRISMA-Guided Systematic Mapping of Sensing, Inspection, Access Robotics, and Condition Monitoring (2016–2026).” Biomimetics, 2026. 505 studies across five functional clusters; identifies limited consolidation into deployable architectures as the principal gap.

Chen et al. “Biomimetic Transition Gait for Quadruped Robot Creeping From Level to Slope Surfaces.” Journal of Field Robotics, 2025.

“A Deployable, Bio-inspired Compliant Leg Design for Enhanced Leaping in Quadruped Robots.” arXiv preprint, 2026. Froghopper-inspired catapult mechanism at prototype stage.

Walas, K. “Legged Robots Beyond Bioinspiration.” Science Robotics 9, eadp1956 (2024).

“The Optimization of Biomimetic Sharkskin Riblet for the Adaptation of Drag Reduction.” Ocean Engineering, vol. 292, 2024. ZnO/PDMS sharkskin surface: up to 12.2% drag reduction and 96.5% reduction in bacterial adhesion in test configuration.

“Marine Drag Reduction of Shark Skin Inspired Riblet Surfaces.” Biosurface and Biotribology, 2017 (foundational review of the marine-application literature).

Standards and reference

US Department of Defense, Office of the Chief Technology Officer. Technology Readiness Assessment Guidebook, February 2025. Definitions of TRL 5–6 (relevant environment) applied in this Signal’s readiness assessments.