note 07 · stoa · 2026
Receipts, not logs: neutrality as a property
An audit trail only counts if the party you would be auditing doesn’t control the auditor. Vendor logs — however detailed — ask you to trust the vendor twice: once to act, once to report. STOA’s answer is structural: every action is canonicalized, hashed (SHA-256), signed (ECDSA P-256) with a key generated on the user’s device, and chained to the previous receipt. The verifier is open source and runs offline; it never calls us. That makes the claim falsifiable by anyone: change one byte of a receipt and verification fails, on your machine, with our servers unplugged.
The subtle part is not the cryptography — it’s the neutrality invariant: the thing that checks the system must not be part of the system. Receipts ship in JUWEL today; the live demo on the JUWEL landing signs and verifies in your own browser.
status: shipping · see it live at juwel.ai · verifier: npx @vextlabs/stoa-verifier
note 06 · cip · 2026
Forgetting curves on toy curricula
The question we’re testing: can a learner add capability without erasing what it had? Our continual-learning program (CIP) freezes what exists and grows new capacity for new skills, rather than retraining in place. On arithmetic-scale curricula the approach holds — new skills land, old skills stay. That is a toy-scale result, and we say so: the real-world curriculum test is still ahead of us, and this note will be updated with the outcome whichever way it goes.
status: toy-scale validated · real-world test ahead · full write-up in preparation
note 05 · systems · 2026
Substrate-agnostic by design
We don’t claim a magic model; we claim a magic record. JUWEL routes each job to the substrate that fits it, and the receipt records exactly what ran. That turns “which AI powers this?” from a marketing claim into a field on a signed document. The full argument lives in the browser-OS architecture blueprint.
status: shipped pattern · read the architecture at juwel.ai/architecture.html
