Probably found the most expensive hardware wallet out there, namely, Sealer2100 currently priced at
$899.
To me, it seems somewhat irrational to pay an inflated price for a device from an obscure manufacturer - one that hasn't specialized in these products before (did they even make such hardware wallets previously?) and thus lacks the extensive experience and engineering track record of established brands - simply because of an eye scanner; after all, even Coldcard has had vulnerabilities.
This is the fresh product to be launched by HPX in 2026.
And like any new product, it could contain numerous bugs and vulnerabilities. I certainly wouldn't store assets in it - at least not initially.
What distinguishes it from other devices is the iris scan, which is used to unlock and authorize wallet and also at signing transactions. The relevant diagrams can be found
here.
I believe the primary function of these HW devices is to protect assets from online (that is, those who lack the physical ability to interact with this device) attackers (something Coldcard failed to do) and from "stupid" offline thieves (via PIN protection). However, these devices can't protect against robbers; if someone confronts you with a $5 wrench, no PIN or iris scanner can prevent unauthorized access to your crypto assets. In a worst-case scenario, robbers could simply hold the scanner up to the owner's eyes to unlock the wallet.
In other words, this distinction doesn't make hardware wallet superior to others (in terms of security).
Their marketing claims it is open-source, but the
Sealer2100 repo describes itself as "auditable but not reusable," so a better term here would be "source-verifiable."
So, it turns out this open source isn't exactly open source?
"The device stores private keys in a CC EAL6+ certified TMC THD89 secure element, physically isolated from the main processor. Iris templates are encrypted and stored within that same chip, with no network transmission pathway. The iris recognition system uses 2048 encrypted biometric feature points with infrared liveness detection, designed to counter spoofing attempts using photographs, silicone replicas, or 3D-printed models." (
https://www.sealer2100.com/en/blog/article?id=16)
And all of this becomes useless if there is even a single vulnerability. What is the likelihood of that on a fresh device?
