A confidential workload that spans a CPU enclave and a GPU produces attestation evidence at every hop, and deployment guides call the result "attested." But if you decompose what the reports contain you're left with grey areas: what was measured versus what executes, when the evidence
Modeling agent messages across ten-node relay networks pushed OpenVTC to adopt the Trust Spanning Protocol and to restructure its stack so the transport question stays routine. Glenn Gore walks through the decision, and DIDComm's Sam Curren explains the design. When the OpenVTC project modeled how its agent
Weeks before Ironwood, the upgrade meant to make its supply checkable again, Zcash cannot prove a four-year-old bug in its largest shielded pool was never exploited. That sounds like an argument against privacy, but the systems it is measured against fail the same way, at greater cost. The
Silicon, math, and a quorum of nodes all ship under the same label. The phrase sets your threat model. On 22 June, two companies shipped products they both called confidential computing. Anjuna, a Palo Alto enterprise vendor, announced Overwatch, a governance layer for AI agents that runs attested inside a
How the unlinkability requirement is being engineered, and the layered cryptographic schemes that may carry it into production Anything deployed today has a shelf life of six to seven years before you have to swap it out for a post-quantum state. That's a short time to amortize.
In barely a year, the industry has agreed on how AI agents should find one another and hand off work. It has reached no such agreement on the harder question as to how an agent proves what it is allowed to do. One of decentralized identity's architects argues
An attestation can prove an enclave is genuine and unaltered since the moment it booted. That's great, but it cannot prove the code inside matches the source you reviewed. Caution, the platform spun out of the security firm Distrust, is built on the wager that providing an attestation
pERC20, EIP-8182, and STRK20 arrived inside three months, alongside ERC-7984, the incumbent. They are four answers to different questions and the people building them are arguing. Within a single quarter, the EVM world produced four separate ways to hold and move a token without publishing the balance. In
The standards for the EU digital identity wallet are largely set. Markus Sabadello argues the framework chose the older, established options over the ones the decentralized-identity community built, and while these choices were practical, they were never merely technical, and this matters because each shaped who ends up in
Harvard's Applied Social Media Lab built Keyring, a reference wallet for the decentralized trust standards taking shape at the Linux Foundation, to make decentralized trust usable. Its engineers believe the social layer of identity needs a graduated alternative to binary KYC and a user experience that catches up
Soda Labs co-founder Avishay Yanai on garbled circuits, the FHE comparison, and what MPC privacy requires The business (aka smart contract) owner decides what privacy looks like, and who's allowed to see what. Avishay Yanai Co-Founder & CEO Soda Labs When I wrote about the three
A court froze funds inside a confidential USDC wrapper then reversed it three days later On May 30, Circle blacklisted the Ethereum contract behind Zama's confidential USDC, a token called cUSDC. It froze the contract, and the contract held the USDC backing every person who had wrapped into
The first deployment of decentralized identity in an open-source supply chain context, and a real-world test of the architectural bet that humans and AI agents belong on the same trust substrate. On building identity infrastructure for both humans and AI agents: We don't see it as
Two acquisitions in eight days show privacy companies buying the layer that lets institutions in. Within eight days in May, two of the more serious companies in onchain privacy made acquisitions that, read together, suggest something about where the field is going. On May 20, Zama acquired TokenOps, a token-
The 47th IEEE Symposium on Security and Privacy wrapped in San Francisco this week. A surface-level scan of the accepted papers list would seem to claim that confidential computing took another beating. Three independent groups published memory-bus and chip-interconnect attacks against Intel TDX and AMD SEV-SNP.
Engineering
Vitalik Buterin published a long essay on formal verification today. It argued that pairing AI-generated code with machine-checked mathematical proofs is the right response to AI lowering the cost of finding exploits. The gist is that Lean is the language of trust, and ZK proof systems and post-
A new Ethereum proposal landed recently that, together with the architecture it depends on, describes a path to L1-native privacy primitives. But several companies have spent years building privacy infrastructure before this path was articulated. What does this EIP mean for the work already done? The proposal EIP-8250
What "private" actually means when you send a prompt to an AI provider On May 6, Canada's Privacy Commissioner published the results of a joint investigation with the verdict that OpenAI violated federal privacy law and three provincial privacy regimes in the development and deployment of
Verifiable AI
What hardware-rooted trust buys an enterprise AI customer in 2026 On May 1, the Pentagon agreed to deploy commercial AI from eight technology firms on its IL6 and IL7 classified networks. The architectural decision sitting underneath that procurement is the same one Yma Health signed in October when it
Verifiable AI
The dominant industry framing for what is happening in the agent economy is that the bottleneck is identity and "Know Your Agent" (KYA) is the AI equivalent of KYC. The frame is correct as far as it goes, but it also collapses what is actually a set of
Identity & selective disclosure
Google used a ZK proof to disclose a quantum cryptanalysis result without publishing the exploit. Trail of Bits forged a "better" proof two weeks later. The episode is a stress test of a new disclosure norm, and the technology is stronger for it. An Early Field Test On
Verifiable AI
A free iOS SDK, in-camera signing on flagship cameras, default-on C2PA on Pixel 10, and a benchmark showing detection breaks under trivial perturbations. The building blocks of Succinct's "prove what's real" media stack are coming together. From Detection to Proof In late
Confidential finance & markets
Every major custodian, market maker, and stablecoin issuer evaluating public-blockchain rails this year is implicitly choosing among three architectures for confidentiality: operator-controlled environments anchored to a public chain (Tempo Zones), client-side zero-knowledge proofs (Aztec), and fully homomorphic encryption (Zama). Each forces a different conversation with regulators,