L2BEAT is an open-source, public-good analytics and research platform dedicated to L2 scaling solutions 💗
Layer 2
Joined July 2021
- Tweets3.4K
- Following234
- Followers42.3K
- Likes2.9K
Most DAO voting happens in public - you can look up a wallet and see exactly how it voted, which allows for bribery and voter pressure. In real-world elections, the secret ballot is an internationally recognised standard that exists to prevent exactly that.
@theInterfold CRISP tries to solve this for onchain voting by keeping individual votes private while making the final results public.
We've added it to the L2BEAT Privacy dashboard. Here's how it works and who you need to trust 👇
Here’s the summary:
Individual votes are hidden from public and network observers.
A chain analyst may still be able to infer some votes, and a future adversary could potentially reveal them. Privileged insiders have a more direct path.
@theInterfold CRISP is live on mainnet, although it's still early. The operator set is in its initial stage, and there have only been a handful of votes so far.
Full breakdown: l2beat.com/privacy/projects/…
The temperature check opened on October 15. Passing it would not activate the system, as implementation would still require technical review, audit, testing and a separate constitutional onchain vote.
snapshot.box/#/s:arbitrumfou…
While delegates consider this additional recovery route, you can explore how Arbitrum governance works today on L2BEAT, including the upgrade paths, permissions and contracts that determine who can change the system.
l2beat.com/layer2s/projects/…
Would you like privacy with frontier AI models? Yes? Well, right now it’s impossible, because we don’t know how.
But we can try to get closer, which is what @ethereumfndn and @OpenAnonProject have been trying to do by co-developing zkAPI!
zkAPI is now listed on our Privacy page. It is a service that offers short-lived OpenRouter keys for ETH through an intermediary called Open Anonymity. You deposit into a shared pool, but your API key request to them contains only a proof that you have deposited, unlinking your specific deposit from the key.
xitter.cf/ethereumfndn/status/21…
And for the dreamers: imagine actual private inference, where neither the model provider nor the operator can read your prompts.
The closest we can get today is TEEs like TDX or AMD SEV-SNP combined with GPUs in confidential computing mode. But that only works for open-weight models and has documented attacks like tee.fail.
In some cases, you can also try to normalize your prompts, like @VitalikButerin did here: xitter.cf/vitalikbuterin/status/…
Wrapped zkAPI + Tor + extra cover traffic + local LLM prompt rewriting in an open-AI compatible daemon, and made a "paranoid mode" wrapper for my agent harness (pi):
Each conversation gets assigned a separate Tor circuit, your requests are masked with random cover traffic, zkAPI ensures you don't leak your identity through payments, and all prompts are locally rewritten (bring-your-own-local-model; I'm using qwen3.8-flash-next)
github.com/vbuterin/zkapi/tr…
My agent harness is wrapped with a bwrap sandbox (also experimenting with MicroVM) and wrapped again with torsocks, so everything it does goes through Tor.
Here's what me asking it to do a simple research task looks like.
It's getting quite usable! Though you definitely still pay quite a bit in increased latency, especially when starting a new conversation.
Also, Tor is far from optimal here for privacy; the cover traffic mechanism is okay against passive adversaries but much weaker if the LLM provider is colluding with the guard and actively manipulating timing. This kind of task really requires a more specialized mixnet, probably take the tradeoff of a higher bandwidth multiplier and built-in constant cover traffic.
Also also, much cheaper zkAPI deposit and withdraw gas costs are coming to mainnet soon.
