Archive
18 articles published in this period.
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.
Bond-and-slash verifiable AI assumes someone audits. But checking costs money and pays nothing, so a rational verifier won't — the verifier's dilemma. Proof of Sampling fixes it with a bounty, and the math says honest inference holds at a 0.74% sampling rate where nobody ever arbitrates.
An on-chain agent acts on facts it reads from an RPC it has to trust. ZK coprocessors swap that trusted read for a proof — and the punchline is succinctness: proving a query over a year of blocks verifies for the same flat ~300k gas as one slot. The mechanism, the anchor problem, real-time proving.
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.
An on-chain agent reports its work in prose — and some of those sentences are fabricated tool calls. zkML proves the wrong thing at minutes per query. The blockchain already wrote an unforgeable receipt for the actions that matter, and a 12ms signed receipt covers the rest.
Frontier open models are now Mixture-of-Experts — cheap to serve in a datacenter, nearly impossible across the internet. The all-to-all expert shuffle moves ~9.5 MiB/token and caps decode at ~67 tok/s even on 400G InfiniBand; on internet links it falls to seconds per token.
Speculative decoding lets a small, even adversarial, drafter propose tokens that a big model verifies in parallel — and the output is provably the big model's, unchanged. Real drafters land at α ≈ 0.6–0.8 for a ~2× win. But the verifier still runs every token, so it's a latency win, not a trust win.
Train a model on the web the last models filled with output and quality decays — the tails go first, then the middle. Provenance is sold as the fix, but a chain proves who signed a blob and when, not that it's human or clean. Signed is not clean.
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.
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.
On-chain data markets promise to pay you for your data's value AND protect your privacy. Differential privacy is the only rigorous knob for the second — and one budget ε can't serve both. The utility collapse, the composition trap, and why no chain can verify the ε you were promised.
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.