"The next internet won’t just be hardware and software – it’ll need trustware." As AI agents become more autonomous, the question of how they verify actions, identity and claims becomes fundamental. Unlike humans, agents can’t rely on intuition, tone or social cues. They need verifiability – and that’s where distributed ledgers come in. The thought piece below from Anand Iyer covers the intricacy and nuance of agentic AI interactions excellently ⬇️ 💡 "Civilizations scale because humans are capable of implicit trust, ai agents are not; their only path to build an agentic society is through a ledger of truth..aka a blockchain. Money is always the first thing written on a ledger as it’s the simplest form of truth to preserve: > Who has what. > Who paid whom. The first step is that agents need a way to exchange value, but that’s only chapter one. Humans rely on memory, reputations and unspoken social cues to decide who to trust, but agents have none of that so every action must be verified, every claim must be proven. Agents need a shared registry where they can anchor identity, reputation and validation. This has a minimal on-chain footprint, but enough to let machines discover each other and know who they are dealing with. Once you have this, the roadmap is clear: 1. Payments first, because money is the most legible record. 2. Identity and reputation next so agents can transact with each other with confidence. 3. Coordination and contracts that enable marketplaces and collective action amongst agents. Why blockchains, and not just any distributed system? Because blockchains are designed for adversarial environments where: > Every record is verifiable > Every state transition is transparent > Consensus emerges without requiring participants to pre-trust each other. Databases can replicate data, and permissioned systems can share logs, but none of those protect against agents lying, censoring or rewriting history. Only a blockchain offers credible neutrality where no one single actor has the keys to the ledger, no one can unilaterally edit the past. For us humans, that distinction may feel subtle because we carry implicit trust into most interactions. But for agents, it’s existential because when one machine interacts with another, it has no body language to read, no reputation to intuit, no fallback court system. Its entire confidence comes from verifiability and blockchains are built for exactly that: a public, append-only memory where identity can be anchored and trust can be established. The agentic internet will run on verifiable truth and money is only the opening act. The endgame is a civilization of agents that communicate, trust and share resources through distributed ledger technology. We have hardware, we have software, now its time for trustware." 🤖 ⛓️ 🔐
"Trustware: The Next Internet Component for AI Agents"
More Relevant Posts
-
🔒Privacy’s Bull Run: Web3’s 2025 Architectural Revolution Privacy in blockchain? Once the underdog, now the backbone. Forget mixers or sanctions drama 2025 is about programmable, composable, invisible privacy reshaping DeFi, DAOs, and institutional flows. Web3 was never meant to be a glass cage transparency audits, sure; strategy leaks? Disaster. 🧩The Privacy Stack Is Live Threshold Encryption : Distributed keys, 50-70% faster multisigs. No collusion, pure discretion. FHE : Compute on encrypted data 20+ TPS on Ethereum, 1000x cheaper via GPU accel. Private lending, no leaks. Recursive ZK & VDFs: Compress proofs, enable private auctions. MEV? Dead. Adaptor Signatures: Scriptless swaps for dark pools bids hidden, settlements public. ⚡What’s Breaking Now Zama’s FHEVM: Launched Oct 2025, enabling confidential Solidity dApps. CZ on X: “Aster’s privacy smokes transparent order books.” Inco’s fhEVM: Cosmos-native encrypted state, no relays. Private randomness for gaming/NFT drops. EU’s EDPB Rules (April 2025): Mandates privacy-by-design; DID for KYC without surveillance. SEC PoS Clarity (May 2025): Staking ≠ securities with ZK-privacy, unlocking institutional PoS. Visa’s VTAP (Q1 2025): FHE-backed tokenized RWAs. Banks like HSBC test private baskets; Gucci joins crypto payments. AI-Blockchain Surge: Partisia’s MPC secures pharma data; Polygon/Eth scale ZK for private AI training. Market projected at $703M by EOY. 🛡️Compliance: Privacy’s New Best Friend Regulators are flipping the script—privacy tech is now a compliance superpower: EU’s MiCA (Live 2025): Requires privacy opts for stablecoins; ZK/FHE for “selective disclosure” is gold. US GENIUS Act (June 2025): Funds ZK for “civil liberties tech,” greenlighting private custody for gov-held crypto. Global AML/KYC Shift: BIS’s 2025 guidelines endorse DID and ZK for auditable privacy—99.9% hidden, 0.1% revealable via quorum. No surveillance needed. Singapore’s MAS (Q2 2025): Approves FHE for tokenized securities, citing “privacy as risk mitigation.” 🧠 Why It Matters Privacy isn’t a feature it’s the foundation for scalable, investable Web3. Discretion drives the next wave of capital, users, and liquidity: Private RWAs: Confidential tokenization for real estate, art, bonds. DeFi 2.0: Dark pools, no-MEV DEXs, encrypted stablecoin rails. Institutional Onboarding: JPMorgan’s Onyx ($1B+/day) and Broadridge’s DLR ($1T+/month) lean on ZK/FHE. Sovereign Data: EU’s GENIUS Act (June 2025) funds ZK for “civil liberties tech.” Privacy isn’t anonymity it’s control. Choose what to reveal, when. Scalability won 2018; privacy owns 2025. Early builders are shaping the decade; laggards are scrambling. Invisible infrapipes you don’t see but can’t live without wins. #BlockchainPrivacy #ZKP #FHE #Web3Infra #DeFi #Crypto2025 #OnchainPrivacy #ZeroKnowledge #CryptoInfrastructure 💬 DM me if you're exploring this thesis we’re already building what comes next.
-
-
CEO Web3 Digital Wallet (wallets AltMe &Talao), Entrepreneur. "All roads lead to wallets"
🚀 Introducing https://VC-Registry.com — the open catalog for SD-JWT Verifiable Credential Types (VCTs) At Web3 Digital Wallet, publisher of Talao and Altme, we first built this tool to improve our own issuers. Today, we’re opening it to the entire ecosystem. 👉 What is SD-JWT VC? It’s a new IETF standard for Selective Disclosure JWT-based Verifiable Credentials, where holders can disclose only the claims they choose, while issuers and verifiers rely on a common definition of credential type (vct). 👉 What is a VC Type (VCT)? It’s the vct value inside an SD-JWT VC: a collision-resistant identifier that tells wallets and verifiers what kind of credential they’re handling. The metadata linked to a VCT includes: - JSON Schema – defines required & optional claims. - Multilingual display[] – ensures consistent labels & descriptions across 30+ languages. - Claim metadata – rendering & selective disclosure hints. 👉 Why does VCT metadata matter? Interoperability: wallets can process credentials from any compliant issuer, eliminating one-off integrations. - Consistent UX: multilingual display[] makes sure names and fields look the same across wallets & issuers. - Reliable validation: schemas ensure issuers and verifiers follow the same structure. - Future-proofing: integrity hashes and stable URLs protect ecosystems over time. - Privacy by design: neutral identifiers prevent leakage while still enabling metadata lookup. 💡 https://VC-Registry.com is more than a registry: 🔹 Generate VC Types from an issuer (OIDC4VCI), from a JSON Schema, or from scratch with AI. 🔹 Localize once in 30+ languages and reuse everywhere. 🔹 Publish with a stable URL & integrity protection, or keep private. 🔹 Discover & reuse existing types to avoid duplication. Built on and fully aligned with the IETF draft: 👉 SD-JWT-based Verifiable Credentials (SD-JWT VC) 🌍 Explore the registry: https://vc-registry.com We’re proud to contribute this building block to the decentralized identity ecosystem. #SDJWT #OIDC4VC #EUDIW European Blockchain Services Infrastructure (EBSI) Potential EU Digital Wallet Consortium (EWC) France Identité FIDES - Accelerating Digital Trust ACN - Alliance pour la Confiance Numérique Compellio Werify.eu Daniel Fett Oliver Terbu OpenWallet Foundation OpenID Foundation Altme Talao
-
Author of 5 IETF Internet-Drafts REM, RRP, RSP, RBIP, RGIP | U.S. Army Veteran | Global AI Recognition | Generative AI, Prompt Engineering Specialist | Innovator in Digital Integrity & AI Transparency
AI is only as trustworthy as its data. Right now, we don’t have a universal way to prove where a dataset came from, how a model was trained, or if an inference was tampered with. That’s why I authored the Sentinel Protocol (RSP), now an active Internet-Draft with the Internet Engineering Task Force (IETF). ✅ What it does: Anchors AI datasets, models, and outputs to blockchain + cryptographic signatures. Provides an audit trail anyone can independently verify. Works alongside existing standards like COSE, CMS, C2PA, and SLSA. Creates a permanent, tamper-evident chain of custody for AI systems. 🌍 Why it matters: From deepfake detection to regulatory compliance, the world is demanding provable AI integrity. Sentinel gives us the protocol to back that demand with math, not marketing. 📄 Full IETF draft here: https://lnkd.in/eNtfvbxY This isn’t a concept. It’s already logged in the standards pipeline. A step toward making AI trustworthy for decades ahead. #IETF #AI #Provenance #Standards #Blockchain #Cybersecurity #Innovation #cxo
-
⚡ 𝗖𝗼𝗺𝗽𝘂𝘁𝗲 𝗜𝘀 𝘁𝗵𝗲 𝗡𝗲𝘄 𝗖𝘂𝗿𝗿𝗲𝗻𝗰𝘆 Every few years, technology reinvents the language of value. We used to call it labor. Then capital. Then cloud credits. Now we call it compute. But we still do not redeem it. We spend energy, spin models, mint hallucinations, and celebrate them as innovation. We pay for the attempt, not the truth. And while the world talks about efficiency, the digital economy continues to reward consumption over completion. We pour terawatts into inference, store petabytes of synthetic text, and call that intelligence. It is not. It is entropy disguised as progress. At BuildETH last week at the Edge & Node House of Web3 in the Presidio, that contradiction sat at the center of nearly every session. Developers, founders, and researchers described new financial rails for digital assets, new ways to bridge crypto, AI, and trust. Beneath the code and vocabulary was the same old tension: How do we prove that digital work was real? How do we measure execution rather than excitement? How do we create an economy that rewards verified contribution instead of performative activity? That question points to a new idea: 𝗥𝗲𝗱𝗲𝗺𝗽𝘁𝗶𝗼𝗻 𝗼𝗻 𝗖𝗼𝗺𝗽𝘂𝘁𝗲 It reframes value as verified execution. The watt-hour, the GPU cycle, the attested action that actually occurred becomes the atomic unit of trust. Proof of Execution replaces belief with measurement. Stablecoins become the neutral settlement layer. Identity and access systems trace the source of work. Reputation emerges from redeemed computation, not from visibility or hype. When that shift happens, AI stops being theater and becomes an economy of proof. It is not about who can raise the most capital or ship the largest model. It is about who can verify results, attribute cost, and demonstrate integrity in execution. Every agent interaction, every retrieval path, every output must carry a receipt that says the work was real. In this emerging structure, APIs do not disappear. They evolve into neutral conduits between agents, retrieval systems, and compute markets. Bitcoin Layer 2 or any neutral substrate can anchor trust for agent-to-agent exchange without owning the network. Stablecoins provide liquidity across those exchanges. Trusted execution environments such as SGX and TDX ensure that what happens inside the processor can be verified outside of it. Redemption on compute is how we stop pretending compute is infinite and start treating it like time. Each watt, each cycle, each attested execution becomes a real-world asset that can be measured, proven, and redeemed. This is where AI, crypto, and sustainability converge. It is where energy meets truth and where digital systems reconnect with physics. We are not just building new AI economies. We are rebuilding the idea of value itself. #AIInfrastructure #BuildETH #ProofOfExecution #Web3 #CryptoEconomics #RWA #DecentralizedCompute #DataSovereignty #Blockchain #AIAgents #DigitalTrust
-
-
Hello Everyone ! The most valuable assets in the digital world are shifting from money to intelligence. As we integrate AI deeper into our products, we're creating a massive new attack surface that demands a new security paradigm. The good news? We've already solved many of these problems. The rigorous, protocol-level thinking from blockchain and cryptography Secure Multi-Party Computation (SMPC), threshold signatures, and secure key management is directly applicable to building AI systems that are not just smart, but also private, verifiable, and secure. In my latest Medium article, I break down the practical application of these techniques: ➡️ How SMPC (the tech behind advanced non-custodial wallets) enables private model training on data no single party can see. ➡️ Using Threshold Signatures to gate model inference, preventing unilateral access to sensitive AI. ➡️ Critical code review patterns to spot cryptographic misuse in AI codebases (hint: say no to random for secrets!). ➡️ Architecting secure key management for model weights, treating them with the same care as crypto wallet keys. If you're working on production AI systems, managing ML infrastructure, or are curious about the future of AI security, this one's for you. Read the full deep-dive here: https://lnkd.in/gBdJixNw The convergence of AI and cryptography is one of the most exciting frontiers in tech right now. I'd love to hear your thoughts: What's the biggest security challenge you're facing with AI in production? Are you exploring ZKPs, SMPC, or Homomorphic Encryption in your projects? #AI #AIsecurity #Cryptography #ZeroKnowledge #SMPC #Blockchain #Rust #MachineLearning #DataPrivacy #CyberSecurity #Tech #Innovation
-
PhD in organizational science, Financial Broker, IPOs Manager, Tax Advisor, Writer, Economic Strategist/ Economist | Advisor | Strategist | Consultant | Researcher | Scholar| Academic| Visionary| Educator
The Regulatory Dilemma: AI, Blockchain, and the Future of Securities Oversight Technological change in financial markets is no longer incremental — it is transformational. Artificial intelligence, blockchain, and tokenization are reshaping how capital is raised, traded, and allocated. For regulators, particularly securities commissions, this creates a dilemma: the tools designed to safeguard markets were built for a world that is rapidly disappearing. Most securities regulations rest on definitions drafted in the pre-digital era. Concepts such as “exchange,” “broker,” or “security” were never meant to capture DAOs, algorithmic trading systems, or tokenized claims. Activity now occurs without intermediaries, across borders, and in real time — often outside the reach of traditional oversight. AI-driven trading and advisory systems promise efficiency but pose transparency and accountability challenges. Blockchain assets democratize access to finance but expose retail investors to fraud, volatility, and asymmetric information. Enforcement becomes complex when no clear party is accountable, raising fundamental questions about the scope of public oversight. Tokenized assets can reduce frictions but introduce systemic vulnerabilities. Regulators operate nationally while risks are global, creating a mismatch between authority and exposure. The most immediate challenge is institutional: regulators must develop expertise to monitor technologies evolving faster than bureaucracies. Without investment in skills and supervisory technology, authorities risk falling behind the markets they oversee. The core policy question is not whether AI and blockchain will transform markets — they already have — but how regulators balance innovation with investor protection and systemic stability. Regulatory sandboxes, international cooperation, principle-based rules, and SupTech solutions will be key. The coming decade will test whether securities regulation, born in the analog age, can reinvent itself for the digital frontier.
-
-
Anyone remember that Black Mirror episode "Joan is Awful"? Basically this crazy ahh AI model created fake content about a real person and it ruined her life. This was a really good episode to watch but the problem is that it's actually starting to happen right now. Deepfakes are everywhere! News organizations accidentally publish fake images. Legal systems can't trust digital evidence and social media has been flooded with fake and manipulated content. Bro our own president has reposted and tweeted fake images 🤦♀️ Basically current detection tools are always playing catch-up with improving AI. So this project aimed to stop deepfakes at the source. Instead of trying to detect deepfakes after they're created, we (Hope R was with me at the EasyA hackathon 😆) built nim8us. Not only is the project name a Futurama reference, it's the name of a system that proves provenance using the blockchain. Basically content creators certify authentic images at creation time -> Anyone can verify authenticity with simple drag-and-drop -> Blockchain provides tamper-proof certificates that can't be faked -> Results show instantly: Verified, Modified, Revoked, or Unknown In terms of tech I used Flare's enshrined FTSO oracles for stable USD pricing with XRPL's settlement layer for tradeable authenticity certificates. Which is super cool because it creates programmable liquidity around verified content (currently impossible on traditional platforms). The business case is massive. We're talking about a $12B+ AI-generated fraud problem that affects all walks of life from journalism to legal systems to social media. When authentic content becomes economically valuable through our system, fake content becomes worthless! So although Joan couldn't prove what was real about her life, I built a system that means you can. Let me know if you want to check out the code or something!
More from this author
Explore content categories
- Career
- Productivity
- Finance
- Soft Skills & Emotional Intelligence
- Project Management
- Education
- Technology
- Leadership
- Ecommerce
- User Experience
- Recruitment & HR
- Customer Experience
- Real Estate
- Marketing
- Sales
- Retail & Merchandising
- Science
- Supply Chain Management
- Future Of Work
- Consulting
- Writing
- Economics
- Artificial Intelligence
- Employee Experience
- Workplace Trends
- Fundraising
- Networking
- Corporate Social Responsibility
- Negotiation
- Communication
- Engineering
- Hospitality & Tourism
- Business Strategy
- Change Management
- Organizational Culture
- Design
- Innovation
- Event Planning
- Training & Development
Helping Companies Maximize the Business Value of Data and AI | ex-CDO advising CDOs at Data4Real | Keynote Speaker & Bestselling Author | Drove Data at Citi, Deutsche Bank, Voya and FINRA
2dI think it would be a good idea to have this for humans too - given many recent and not so recent developments, usual signifiers of trust have eroded greatly.