# Ethereum’s Guarantees Shouldn’t Stop at the Protocol Source: https://x.com/charliestlouis/status/2122358084131574264?a=C_00U73vua5BSwUkn7M&s=46 Kiwi News: https://news.kiwistand.com/stories/Ethereums-Guarantees-Shouldnt-Stop-at-the-Protocol?index=0x6afb90087fed008fcb26ab1e70d93b22919b843f778a022276753f81adceadce1943db42 Submitted by misha on 2026-11-16 UTC, 1 upvote, 0 comments ## Summary Ethereum's guarantees do not automatically extend to the wallets, applications and infrastructure users rely on, and the author argues this deserves more attention as the ecosystem grows. The author writes from the Ethereum Foundation's Access Layer team, particularly App Enablement, which is helping teams adopt tools such as Kohaku for privacy-preserving accounts, transaction assertions, anonymous RPC infrastructure and verifiable frontends. The text cites the BadgerDAO frontend compromise in 2021, the Curve DNS hijack in 2025 and the compromised Safe interface behind Bybit's $1.4B hack that same year as examples of these risks. ## Article Ethereum’s guarantees don’t automatically carry through to the applications and infrastructure people use. And I think that deserves more attention as the ecosystem grows. Following some conversations at Devcon and other recent events around CROPS and EF’s priorities, wanted to share a bit of perspective on what we’re working on within the Ethereum Foundation’s Access Layer team, particularly App Enablement. Most users interact with Ethereum through wallets, applications, transaction infrastructure, and now even LLMs. Those layers can reintroduce intermediaries, concentrated security risks, and points of control, even when the underlying protocol is decentralized. A few practical examples: Why should a single frontend or RPC provider going down prevent people from accessing DeFi when Ethereum itself is still running? Why can smart contracts undergo audits and formal verification while users have no easy way to verify that the app running in their browser is the code developers published? We've seen the consequences with BadgerDAO's frontend compromise in 2021, Curve's DNS hijack in 2025, and the compromised Safe interface behind Bybit's $1.4B hack that same year. And why should checking your balances require revealing your addresses and activity to an RPC provider? These are real problems today, not just concerns for people who identify with a particular set of principles. Every intermediary introduces a potential point of trust, control, or failure. Removing unnecessary intermediaries can reduce concentrated risk, but only if the alternatives are secure and usable. That’s a big part of what we’re working toward in App Enablement. We work with teams to understand their constraints, help them explore and adopt existing infrastructure, and bring recurring blockers back to the people working on the protocol. A big part of this is through a CROPS lens: censorship resistance, open source, privacy, and security. I’ve seen criticism that CROPS is an ideological movement most people don’t relate to or need. But I think that misses the point. These properties have practical benefits for everyone, whether or not they care about CROPS. And it’s not all or nothing. Apps can prioritize different properties without satisfying every one of them. There are real technical tradeoffs, particularly while privacy-preserving infrastructure is still maturing. Not every property can be maximized simultaneously, and being on a spectrum doesn’t mean every implementation offers meaningful guarantees. That’s part of building practical technology, not a reason to abandon progress. Teams should be transparent about their protections, limitations, and dependencies. The goal isn’t to force every app or business into a particular model. It’s to make credible, secure alternatives available for users, including paths for self-custody and self-sovereignty. So what does that look like in practice? For wallets and apps, we’re helping teams explore technologies like Kohaku for privacy-preserving accounts, transaction assertions for stronger signing protections, and anonymous RPC infrastructure for private reads. I also think there's an opportunity to advance verifiable frontends, so users can independently check that the application they're interacting with matches the code developers actually published, rather than blindly trusting whatever a website serves them. Users should be able to verify what they’re running and what they’re signing, rather than blindly trusting an intermediary. We’re also looking ahead to PQ signing and wallet integration. The point isn’t to have every team build everything from scratch. It’s to make these capabilities easier to adopt while identifying what’s still missing. I also think EF has a unique opportunity to look beyond what’s commercially important today. Ethereum is still a relatively small ecosystem, where personal relationships and trust in the teams behind applications can go a long way. But that doesn't necessarily scale. If we don't make these properties practical at the application layer, what's stopping us from recreating the same problems we have with traditional finance? Users getting locked out of services, intermediaries controlling access to their money, or switching financial products becoming unnecessarily difficult. These properties matter today, but I expect them to become even more important as Ethereum grows over the next 1–10 years. Building the infrastructure to deliver them securely and at scale takes time, and I think EF is particularly well-positioned to invest in that work, even where immediate commercial incentives are limited. If we wait until everyone needs these properties, it’s already too late. Meanwhile, institutional adoption is accelerating. Organizations like @Etherealize_io and @ethereuminsti are advancing institutional adoption, while @ethlabs_org is supporting Ethereum's development through R&D. Companies like @Sharplink and @BitMNR are also bringing additional capital and attention to Ethereum. That’s a good thing, and I want to see much more of it. But institutional adoption shouldn’t stop at holding ETH or accessing it through centralized intermediaries. There’s a much bigger opportunity to bring institutional financial activity onchain through stablecoins, DEXs, lending markets, and other DeFi infrastructure. And the opportunity goes beyond making existing financial services more efficient. These systems already demonstrate what’s possible when financial services operate on open, globally accessible networks without relying on a single institution. The next step is making sure users can access and interact with them without reintroducing the same gatekeepers, dependencies, and risks we’re trying to move beyond. Ethereum should have it all. Institutional DeFi, consumer apps, and applications that prioritize different CROPS properties. These aren’t mutually exclusive. Institutions need privacy, security, open standards, and reliable infrastructure too. Censorship resistance and credible neutrality can also reduce dependence on individual gatekeepers and provide more credible alternatives when existing access is restricted. To me, EF doesn’t need to lead every growth initiative. It has an important role in supporting shared infrastructure and work that might otherwise be underfunded or overlooked, while others focus on adoption and bringing more economic activity onchain. We shouldn’t have to choose between bringing the world’s financial activity onchain and preserving the properties that make Ethereum worth building on. The objective is an Ethereum ecosystem where users have meaningful choices, including credible, self-sovereign alternatives when they need them. And if you think that means EF is missing the boat, that’s fine. Ethereum will still be here when you need it. --- For more context on EF’s structure and how the teams fit together: https://blog.ethereum.org/2026/06/23/ef-structure