Opacity Labs
Verified account This account is verified because it's an official organization on X. Learn more@OpacityLabs
Tokenized securities and portable identity.
1K Following 6.3K Followers 2.2K Posts
Latest posts 5
Drink plenty of water today.
This isn't a tokenization or identity post it's just a reminder 🚰
This isn't a tokenization or identity post it's just a reminder 🚰
Tokenization is useful because programmable assets can travel outside of trad venues, yet within the rules set by the issuer.
Opacity is deterministic, neutral tech that allows access to financial assets according to the rules set by the issuers within their legal obligations.
Opacity is deterministic, neutral tech that allows access to financial assets according to the rules set by the issuers within their legal obligations.
What you're describing is internal transfers between accounts at one brokerage, that's been possible since brokerages existed. The blockchain here is doing nothing the broker's own account database doesn't already do.View quoted post on X
Tokenized receipts are fine, wBTC and USDC are good examples. They work because it allows any AMM, Vault, Venue, CEX etc to work with each other without bespoke coordination, they just adopt the same software standards.
What people want from tokenization isn't no rules, it's no gatekeeper. Define the rule(s) in a smart contract and let anyone show up and prove they satisfy the rules. What Brett advocates for is the opposite, the rule is "the wallet must have a brokerage account with Securitize to transact."
Extrapolate the model and the end state is one brokerage everyone has to onboard with, because that's the only place the tokens can go.
I don't understand how anyone calls this innovation, it's rewinding the clock of American capital markets.
Until the input carries provenance of its source, verifiable inference use cases will be limited. Verifiable data comes first.
The bottleneck verifiable AI inference will hit:View quoted post on X
A proof of inference only shows that the model ran correctly. The user supplies the input, so a forged input still yields a valid proof. Verifying the input is authentic is the bottleneck.
@AnthropicAI released FlashProve yesterday: a zkSNARK that an output came from a committed set of model weights. The user still checks only that their provider ran the right model. The moment the proof is shown to a third party as evidence, it is only as good as the input, and the proof says nothing about where the input came from.
Verifiable compute moves the trust from the party running the model to the party supplying the data.
That is why these proofs worked on blockchains first. A rollup's inputs are posted to the L1, so the proof covers data anyone can check. Off chain, there is a chronic shortage of verifiable data. A bank balance, a payroll record, a cap table or a lab result sits on a server that shows it to you over an encrypted connection. Once it leaves that server, it becomes a screenshot or a PDF.
Until the input carries provenance of its source, verifiable inference use cases will be limited. Verifiable data comes first. Verifiable compute is what you do with it.
@OpacityLabs
Attribute-based Credentials attest to facts such as citizenship, accredited investor status, absence from sanctions lists, without revealing the underlying data.
opacitylabs.com/reusable-crede…
opacitylabs.com/reusable-crede…
