Safe Research Manifesto

Safe Research is a new R&D team formed by core contributors from Safe, committed to advancing self-custodial wallets and accounts for Ethereum. We focus on the cypherpunk pillars of security, censorship resistance, and privacy.

Our pledge

We, a small team of core contributors from Safe, declare our commitment to evolving self custody through our newly formed R&D team, Safe Research. We pledge to do applied research for self custodial accounts, focusing on our three core pillars: security, censorship resistance and privacy - pillars which are critical to the maintenance and growth of Ethereum’s cypherpunk soul. We also pledge to transparency—researching in the open and releasing all the technology we build as open-source.

Why self-custody is still not solved

10 years since Ethereum launched, we still get wake up calls everyday that the industry needs to improve its security. Too many wallet and infra-level architectures rely on single points of failure and offchain trust. Self‑custody remains painfully difficult, and most “innovation” is really web2 tooling awkwardly repackaged, We recognise that key building blocks are still missing to make onchain self custody accessible and secure to all crypto users. Our mission is to identify and construct those missing pieces.

Self custody without compromise

In the spirit of the cypherpunk ideology, we refuse to accept the status quo of centralised infrastructure and broken self-custody stacks. Our R&D will be applied to self custodial accounts/wallets, to enable the ecosystem to grow beyond “centralised UI sugar” and prepare for new EVM developments and black swan events. We also recognise that security in Web3 is a constantly evolving challenge, requiring continuous adaptation. As threats become more sophisticated, so must our R&D and defences.

We first specifically target the smart account/wallet layer, as they are the window between a user and the Ethereum world, and a user only benefits from any decentralization, censorship resistance, security, privacy, or other properties that Ethereum and its applications offer to the extent that the wallet itself also has these properties. We want to forge a path where all components of the wallet stack – signing, recovery, storage, queuing, even interfaces – can be run trustlessly by the community.

Our goals

We measure success not just in papers or prototypes, but in adoption and real-world impact – we aim to see our research results integrated into wallets, dApps, and standards across the ecosystem.

Our pillars

Security

As mentioned, security is paramount and needs to improve, we will never compromise on security in our R&D without exception. By security, we mean

• (i) protecting the user from the wallet developer being hacked or proposing malicious transactions

• (ii) protecting the user from their own mistakes.

Censorship resistance

Censorship resistance is vital to realising self-custody’s promise of neutrality and freedom. While smart accounts and wallets can offer uncensorable control thanks to their decentralized foundations, practical demands (performance, cost, UX, and business incentives) too often push users back toward centralised solutions. Our R&D team is dedicated to moving beyond this “centralised UI sugar” to preserve self-custody’s integrity, resilience, and trustlessness.

Privacy

Today, privacy can no longer be ignored. AI is greatly increasing capabilities for centralised data collection and analysis whilst exponentially expanding the scope of data that we share voluntarily. We need to push for a world where the default should be private, as its a fundamental right.

Hence we push for the integration of privacy-preserving technologies into smart accounts and the wider Ethereum ecosystem, to enable users to transact, interact, and authenticate with confidence and anonymity.

The sweet spot

We acknowledge that we have not reached an appropriate baseline of convenience and accessibility in Web3 UX.

• Convenience: Refers to the ease of use for someone already capable of interacting with the system.

• Accessibility: Refers to making the system usable and safe for as many people as possible, including those with no prior knowledge or technical literacy.

These are related as there is a pivot point when convenience becomes so poor that accessibility is no longer possible. Moreover, if you over-optimise on convenience, you will compromise its security.

Self-custody should not be a burden – so we prioritise user experience as highly as security. When it’s too complex, users don’t just walk away – they YOLO through it, blindly signing what they don’t understand.

Thus, we are committed to eliminating the UX hurdles that have historically forced people back into centralised custody. We also understand that there are always tradeoffs when it comes to UX and security. We seek the “sweet spot” where security features (social recovery, security checks, private tx’s and decentralized backups) are implemented in ways that don’t burden or confuse the user.

Join us in building the future

Our manifesto is a promise and a challenge – one we cannot achieve alone. We invite every like-minded cypherpunk to join us in this mission - whether it’s by contributing code, proposing research ideas, or providing feedback on our prototypes.

Reach out to us via research@safe.dev or contribute to the Safe Research stack on GitHub. Let’s make Ethereum’s original cypherpunk vision real, one commit at a time.