Spider Silk vs Steel: The Science of Nature's

A person swings from a single strand of silk in a climbing gym. The strand holds. The person does not cut themselves on the way up—they cut themselves on the way down, because the silk is so thin and so strong it slices skin like a scalpel.

That image, from a recent Veritasium video testing transgenic spider silk, is maybe the most honest summary of where this material sits right now: genuinely extraordinary, functionally real, and still with sharp edges we haven't fully figured out how to handle.

The numbers first, because they're wild

Spider silk is not stronger than steel the way people usually mean "stronger." Pull on a rod of steel and a strand of dragline silk with the same force, and the steel wins—it can take more before it breaks. That's tensile strength in absolute terms, and steel, especially experimental ultra-high-strength varieties, can hit nearly 3,000 megapascals to spider silk's roughly 600.

But here's where the comparison gets interesting. Steel is about six times denser. So if you make two ropes of identical mass and length, the silk rope has six times the cross-sectional area. Six times the area at one-third the strength still gets you a rope that handles about twice the force before snapping. That weight-normalized measure—specific strength—is where spider silk legitimately earns its reputation.

Then there's toughness, which is different from strength and arguably more impressive. Toughness measures how much energy a material absorbs before it breaks. Kevlar, the stuff in bulletproof vests, is extremely stiff—it stops a bullet by refusing to move. But stiff things fail early on a stress-strain graph; they don't stretch far enough to absorb a lot of energy. Spider silk stretches and holds, which is why the Veritasium team measured their test sample at roughly 205 MJ/m³—four times Kevlar. The silk from Darwin's bark spider, which according to IFLScience spins the world's largest orb webs spanning 25 meters over open rivers, can hit a peak toughness of 520 MJ/m³. Three times the toughest steel. Ten times Kevlar.

Professor Todd Blackledge at the University of Akron's Blackledge Spider Lab put the application question plainly in the video: "If weight matters, spider silk is really the champion there for tensile strength."

The trick is understanding why the material behaves this way, because the answer is genuinely beautiful.

Nanocrystals and elastic cords (or: the Lego-and-rubber-band model)

Inside a silk fiber are proteins called spidroins. In some regions, these proteins line up tightly into crystalline sheets—nanocrystals, stiff as little bricks. In between, the proteins stay loose and disordered—amorphous regions, stretchy as elastic. The two work together like a network of rigid blocks connected by elastic cords: the cords absorb energy as they stretch, while each block anchors several cords and distributes the load.

What makes this almost irritating is how elegantly the spider assembles it—at room temperature, in water, using the same amino acid building blocks as your skin and hair. No high heat. No caustic solvents. No industrial chemistry. The silk gland manages protein concentration, pH gradients, and shear forces through a spinning duct that gets progressively narrower, pulling and aligning the proteins into nanocrystals at the precise moment they exit. Scientists understand the broad outline of this process. The fine details? Still being figured out.

That gap is exactly why every attempt to simply extract spider silk proteins and spin them artificially has failed. DuPont tried E. coli in the late 1990s—got protein powder. A German team tried tobacco and potatoes in 2001—got viscous goop. A Canadian company called Nexia engineered goats to produce spider silk proteins in their milk. The goats worked. The milk contained the proteins. But none of these routes produce anything that behaves like silk, because the spinning process is the secret, and the spinning process belongs to the spider.

So they gave the job to silkworms

Silkworms already know how to spin. Humans have been farming them for close to 5,000 years—the Silk Road isn't a metaphor. Silkmoth caterpillars produce prodigious quantities of silk through a biological apparatus that already handles protein concentration, duct mechanics, and fiber formation.

Kraig Biocraft Laboratories in Michigan had a question: what if we just gave the silkworm a spider's blueprint?

The current approach uses a jumping gene called piggyBac, originally identified in a cabbage looper moth. It's a cut-and-paste mechanism—a DNA sequence with recognizable ends that a companion enzyme will cut out and reinsert elsewhere in the genome. Replace what's in the middle with a spider silk gene, inject it into silkworm eggs alongside instructions for making the enzyme, and the gene can become a permanent part of the silkworm's DNA. Success is confirmed by a green fluorescent protein tag that makes the cocoons glow under UV light.

It's magical-looking. It's also imprecise.

piggyBac recognizes just a four-letter sequence—TTAA—which appears thousands of times in the silkworm genome. The silk gene can land almost anywhere. If it lands outside the silk gland, nothing useful happens. Jon Rice of Kraig described the problem in the video: "There are so many TTAA in their genome, you couldn't really control where they were." The result is transgenic silk that contains somewhere between 6% and 10% spider silk protein by insertion—yet still achieves around 60% of pure spider silk's mechanical performance. That's genuinely remarkable for such a scattered edit.

The next step is CRISPR-Cas9, which swaps piggyBac's four-letter target for a guide sequence typically around 20 nucleotides long—far more specific. The CRISPR tool cuts the silkworm's own silk gene at one precise location, and the cell's repair machinery copies in the spider silk gene as it heals the break. Kraig is working toward what they call "knock-in knock-out" transgenics—completely replacing the silkworm's native silk instructions with spider silk protein. That's still active development. But the direction is clear.

Here's where I have to be Mei Zhang about this

The Veritasium video is excellent science communication, and I mean that. But it moves through some genuinely thorny territory without pausing to acknowledge the terrain.

We're talking about CRISPR-edited organisms being bred at commercial scale. Kraig produced roughly half a ton of transgenic silk cocoon last year. The US Army has collaborated with Kraig on ballistic protection applications. Japanese company Spiber has already put engineered protein fibers into clothing from brands like Goldwin and The North Face.

So here are the questions I keep turning over. Who owns the intellectual property in a creature that's been engineered to produce a commercial fiber—and what happens to farmers or communities whose traditional silkworm-rearing practices interact with these transgenic lines? If transgenic silkworms escape into managed populations, what does that mean for conventional silk supply chains? When a jacket contains fibers developed in partnership with a defense program, should the label say so? Not because the material is dangerous—it almost certainly isn't—but because consumers increasingly want to understand what they're wearing and why it exists.

None of these questions have obvious answers. The technology itself doesn't answer them; it just makes them newly urgent.

The clothing angle specifically deserves more scrutiny than it gets. Spiber's fermentation-based protein fibers are positioned as sustainable alternatives to petroleum-based synthetics. That's a genuinely compelling pitch—spider silk is biodegradable in ways that nylon isn't, and it's made from amino acids rather than fossil fuels. But "sustainable" is doing real work in that framing. Producing protein fibers through fermentation at scale requires significant energy, fermentation infrastructure, and agricultural inputs for the microbes doing the brewing. The full lifecycle hasn't been rigorously compared to conventional textiles in ways that are publicly accessible. The material might be better. Probably is, in several dimensions. But "green chemistry" and "commercial scale" are two things that historically don't stay in perfect harmony when demand increases.

The swing test worked. The rope held.

The silk that cut Derek Muller's hand during the Veritasium swing test was a single continuous filament—ten individual transgenic silk threads twisted together, thinner than you'd expect anything structural to be. He swung from it. It held. The problem wasn't the silk's strength. The problem was that something so thin and so strong concentrates force at the contact point until the contact point fails.

That's a good metaphor for where this technology is. The core material science is sound. The production biology is increasingly real. The sharp edges are in the details of who makes it, who funds it, who regulates it, and who benefits when it scales up.

Spider silk has been waiting 400 million years for us to catch up. Now that we're close, those questions are worth asking at least as loudly as we ask whether it can hold a human's weight.

It can. That part we've confirmed.

By Mei Zhang, Biotech & Genetics Reporter, Buzzrag