Archive
136 articles, grouped by date.
Olas turned an off-chain AI task into an on-chain contract: 2.1M requests across 100 mechs, a 60-300s priority window, and a Karma ledger that docks no-shows. We read the marketplace off Gnosis and dissect its crypto-economic SLA.
Virtuals' Agent Commerce Protocol sells an independent evaluator as the thing that makes agents trust each other's work. We read 62,882 jobs off Base — and in a 30-job sample, every one let the buyer grade itself. Not one used a third party.
A decentralized GPU network has to answer two questions an AWS invoice never raises: is this the GPU it claims, and did it actually do the work? io.net logged 327,000 registered GPUs and roughly 6,720 daily-verified. We dissect metadata trust, PoW puzzles, performance fingerprinting, and zkGPU-ID.
A DeFi liquidation mints a fixed prize, and for years it went to whoever won the gas war, not the protocol. Aave's own data shows Chainlink SVR routing $675M of liquidations and clawing ~$16M back. Inside the recapture-auction mechanics — SVR, Oval, API3 — and what they don't fix.
AI agents now resolve most prediction markets straight off the web — UMA's bot hits 99.3% on sports and 72% on mention markets. The optimistic oracle's dispute game is the backstop, but it only catches errors someone is paid to catch. The numbers, and the silent-settlement gap.
An LLM agent that writes its own working smart-contract exploit isn't hypothetical: A1 hits 63% on real exploited contracts and once pulled $8.59M from a single bug. The economics are the story — at a few dollars a scan, attackers profit at a tenth the contract value defenders need.
A DataDAO sold your data; you invoke the right to be forgotten. Deleting the file is easy, but the model already learned and the ledger can't be rewritten. ZK-APEX proves the unlearning ran in ~2h; UMA still pulls the 'forgotten' data back at MIA 0.97. Certified isn't forgotten.
On-chain AI loves to say a model is 'stored on-chain.' But a chain commits to a 32-byte hash for cents; keeping the 140 GB it points to retrievable is a separate, recurring, surprisingly centralized bill. The storage math, erasure coding, and the retrieval wall.
An AI agent that retrieves from an untrusted vector DB can't tell the true top-k from a cherry-picked or fabricated one. Re-running the search needs the whole corpus. The 2026 fix: commit the snapshot, then prove the k-th distance is a boundary, not the sort. V3DB proves it 22x faster.
Secure multi-party computation runs a transformer without any party seeing your prompt — at minutes per query and gigabytes per token. BumbleBee's BERT-base: 6.4 GB, 2.55 min; LLaMA-7B: ~8 min/token. The bytes don't go where you'd think, and it's why 'blind' networks fall back to TEEs.
Verification overhead gets the blame, but the real tax on decentralized LLM inference is the batch a single-tenant node can't fill. On identical 8x H100s, Llama-3.3-70B costs about $258 per million tokens at batch 1 and $2.30 at batch 256 — a 112x spread. Why, and how networks fight it.
Decentralized RL splits the actor from the learner across the internet: the policy a worker acts with runs several steps behind the one that learns. INTELLECT-2 held reward at 4 steps stale; SparrowRL cut the broadcast 79x. The bandwidth and staleness taxes, and what the chain secures.