Replies (6)

I like the concept but how does a user verify the claims of entropy?
Thanks for the great question. Modulo keys are made so you never need to audit a hidden process, which is why we made them. No chip, no algorithm, no firmware generated randomness for you — you are the entropy. Each stainless steel bit has two mechanically indistinguishable orientations, so every insertion you make is a fair coin flip, verifiable by direct inspection: hold the bits, measure them, examine the key, and confirm there is nothing about the hardware that favors one orientation over the other. Unlike a hardware RNG, there is no firmware standing between you and the entropy. The entire entropy process happens in the open. If you want to verify the output yourself, run a few throwaway assemblies, record the bits, and examine them on an offline machine. The NIST SP 800-22 suite is really good and includes valid tests on sequences of 100+ bits — Frequency, Runs, Longest Run of Ones, Cumulative Sums, and Serial among others — even a single 128- or 256-bit output can be checked for bias, drift, and patterning. If you want to go the distance, concatenate several practice runs: thousands of bits characterizing your own insertion habits with genuine statistical power. What you’re testing at that point is if your generation method is random enough. Once satisfied, disassemble and generate your production key fresh. It never touches the interweb, and no one but you has ever seen it. Sorry for the length of the response but it’s a really important question. Don’t trust. Verify. In metal.
Agree that I need to see a video Can’t seem to figure out how they are getting the bits in the slots and then reading the 1s and 0s 🤣 Love the concept though now my question is are the 24 words worth the extra $180 if I am thinking about buying 2 of the minis to create seeds that would be used in a 2 of 3 setup?