This month has made it undeniable, running significant amounts of ethereum:native stake on single points of failure is not a safe long term strategy for your funds. We’re introducing Validator Beat, a simple framework to make sure your stake is operated securely. Stage 1 means no single point of failure gets you slashed. Stage 2 means no single point of failure takes you offline. Stage 0 validator failures could cost you weeks of rewards at best, or decades worth or even everything at worst. Don’t cut corners in a space with asymmetric downside like this, it’s picking pennies up in front of a steamroller.

Why you shouldn't stake your eth on a single point of failure setup in one picture

Most the public internet being encrypted by these 6 “private keys” says a lot about single point of failure setups. Your eth staking bag deserves better. Stake only on stage one distributed validator setups, where no single operator can pick a private key like this for you 🫠

Ethereum is considering correlated downtime penalties for stakers, an idea by @VitalikButerin. The aim is to disincentivise centralised points of failure before they cause a mass slashing event I made a website to understand the EIP and why we should adopt it 🔗 below, tl;dr👇

What are some of the biggest single-private-key wallets that sign programmatically? MEV bots, Exchange hot wallets maybe, not oracles (multisig), not bridges (multisig), not defi, …. Anything with a decent chunk of value to protect matures to using multiple setups with fault tolerance and diversity for security. On the consensus layer, I think we’re lagging behind through complacency. I would say less than 1% of stake if not 0% are running on proper no single point of failure setups, depending on how militant you want to be about how you define it. I think we hand wave at the fact that a hacker can just destroy your money not steal it, so it’s not as compelling for them to go after, which may be true, but I’m not sure it’s enough; when the costs of AI-enabled spear phishing campaigns are decreasing and their quality increasing. Defence needs to be successful every single time, offence just has to succeed once. That’s not symmetrical. We had a near-miss with 10% of the validator set just last year we’ve mostly forgotten about. It’s also rational to accumulate stolen keys rather than coming forward once they’ve stolen some. A hacker’s leverage grows non-linearly with the more ETH they can slash. We don’t know how many keys are already silently stolen, but I don’t think it’s 0. Let’s try and fix this before not after a disaster. Let’s run more distributed validators, and reduce the amount of single points of failure in our staking setups, and make Ethereum more anti-fragile.

@dsrvlabs have been running distributed validators since our very first public test network four years ago. It’s a great pleasure to have them now adopting distributed validators for a significant percent of their Lido curated set stake. This deployment will allow them to run up to 40 0x02 distributed validators as part of Lido’s ongoing consolidation transition.

"In premise, penalizing correlated failures pushes the network toward the diversity we all want, and if the math checks out the lift on the client side is small. We had initial hesitation but the methodology was recently fixed by Oisin at Obol. A client bug that takes a large cohort offline gets penalized extra while minority clients and well-architected operator with diversified infrastructure barely notices. With the updated, clear and data-backed statement of exactly what behavior this is meant to discourage we are interested in seeing this ship if there is time." Glad to see vote of confidence and the Lodestar team in favour of EIP 7716 in Hegota capacity-permitting 🎉

I would say I'm surprised, but really I'm just disappointed with the EF Protocol's continued presumption of decentralisation as a given. Let me be clear: FOCIL doesn't work if the attester set doesn't want it to. The protocol priorities post focuses on post-quantum as a goal, fine, and pays lip service to CROPS; "FOCIL is Hegotá's consensus-layer centerpiece. Our appetite beyond it is close to zero unless an addition directly supports post-quantum readiness." The elephant in the room the EF continues to ignore is that the chain is centralising at a rapid pace, and there is a lack of interest in either acknowledging the problems, or prioritising fixing any of them. For example, correlation penalties (7716) they describe as: "Adds slashing-logic surface for validator correlation with no CROPS or PQ contribution. Not a priority in a fork, we are trying to keep the CL scope light." This is inaccurate to start, 7716 has no slashing logic, and its misguided at best to assert that it doesn't contribute to CROPS. 7716 is the best EIP on the table to improve client diversity and penalise centralisation. It's not a panacea, but its what we have, it should not be DFI'd. Spending a year on H-star to ship FOCIL without addressing the validator set's underlying issues risks the big few centralised staking entities saying 'actually nah' and refusing to attest to slots with ILs they don't like. This costs them no rewards, thinking we'll socially slash them for it is wishful thinking, and now Ethereum is compromised and you need permission to use the chain. Preventing this outcome is not a matter of one EIP on the margin, it needs a cultural shift within the EF to realise they don't have a divine right from God to a decentralised chain, and that they can't just focus on academic challenges like PQ while ignoring real challenges like the ongoing acceleration in chain centralisation. We need quarterly reporting on the state of the chain, to make it harder for people to keep their head in the sand as to the severity of the situation. We need at least one EIP per fork focused on improving the nakamoto, staking gini, or other aspect of the validator set, and we need pragmatic, action-oriented leadership who care about Ethereum being CROPS beyond just a purity test. (I'm hoping that's @pcaversaccio). There are some people in the EF that care about the chain being CROPS for real, but unless we the community are outspoken and steadfast in our demands, they won't be listened to. Please show them why we're supporters of Ethereum and not any another centralised chain, please advocate for decentralisation and CROPS in H-star and beyond. Thank you.

The amount of vulnerability reports we get has also followed a similar graph over the same time period, it’s been hard to deal with. Lots of duplicate won’t-fix type issues (yes you can read our public APIs, that’s what they are for) but many valid modest issues that we’ve addressed and paid out for. Almost all of Ethereum’s $100b of active stake has at least one single point of failure that could leak its keys and get it slashed. Bad OpSec, bad supply chains, high exposure to particular software, operating systems, CPUs, and more could lead delegators who trust solo setups to up to 100% loss of their principal. We’re going to help simplify and highlight these single points of failure in a new initiative aimed at improving Ethereum’s security and its stakers’ ETH safety. If successful, most organisations will move from single point of failure setups to staking systems without a single point of failure, such as multi-operator distributed validator setups. This will better secure their stakers’ funds and ensure Ethereum never goes offline.

This is a massive breakthrough for post quantum blockchains. Threshold signatures are a fundamental building block for the secure use of a blockchain and until now, there was no solution for how to make them with a post quantum scheme. Ethereum will be post-quantum secure and it won’t have to give up application and consensus level security to achieve it. Post quantum distributed validators when?