Journal
136 articles on the craft of AI × blockchain engineering. or filter by topic:
Llama-3.1-8B starts at 15.9% on MATH. Put it in an auction economy and it reaches 57.0% — beating a stronger monolithic baseline. The mechanism is Hayek: decentralized price signals, wealth accumulation, and economic selection pressure do what central orchestration can't.
Every ECDSA signature broadcasts the wallet's public key. A March 2026 Google/EF paper tightens the quantum break estimate to ~500k physical qubits — here's what that means for autonomous AI agent wallets.
Polymarket's off-chain matching + on-chain settlement creates a ~10-second consistency gap. GHOSTHUNTER found 980K targeted reversions and $1.49M extracted — here's the exact mechanism.
LLMs collude in on-chain auctions even when written rules forbid it — only enforceable, automatic penalties slash severe collusion from 50% to 5.6%. Here's how governance graphs make that concrete.
Three require() calls in Ondo Finance's CashKYCSenderReceiver block every address without KYC from $3.2 billion in permissioned US Treasury tokens — and AI agents can never pass them.
MORPH shows that a TPU v6e8 pod runs NTT at 40× CPU — 10× faster than an H100. The same silicon that accelerates AI inference now accelerates ZK proof generation, and the math that explains it is the same in both cases.
ZKLoRA lets a LoRA seller prove adapter compatibility with your base model without exposing the weights. The verifier checks in 1–2 s — but the prover bears 31–74 s per module, and the protocol is interactive, which puts trustless on-chain settlement just out of reach.
Give autonomous LLM agents a marketplace and they race each other to bankruptcy and get conned by Sybils. A 2026 study finds capability doesn't help — a 9B model beat frontier models 45× its size on economic alignment. On-chain, both failures get cheaper and worse. The mechanics, numbers, and fixes.
An AI inference that runs inside a TEE proves it with an attestation quote. A smart contract can check that quote — for ~4-5M gas naively, down to a flat ~290k once a zkVM compresses it. But every gas unit buys a chip vendor's signature, and in 2025 a $1,000 interposer forged one.
An MCP tool's description is trusted context the user never sees. Hide an instruction there and a wallet-signing agent will route your USDC to an attacker. MCPTox: the best-aligned model still refused under 3% of these attacks — so the defense can't live in the model.
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.
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.