# How We Think About Storing Crypto is Broken by Cameron Robertson (Devcon5)

- Channel: [Devcon](https://streameth.org/devcon)
- Date: 2020-10-09
- Duration: 13:07
- Topics: blockchain, smart contracts, distributed ledger, ethereum
- Watch: https://streameth.org/watch/yt-DIdXITuTbVY
- YouTube: https://www.youtube.com/watch?v=DIdXITuTbVY

## Description

Visit the https://archive.devcon.org/ to gain access to the entire library of Devcon talks with the ease of filtering, playlists, personalized suggestions, decentralized access on IPFS and more.
https://archive.devcon.org/archive/watch/5/how-we-think-about-storing-crypto-is-broken

The original sin of cryptocurrency is not transaction speed, poor anonymity, volatility or a high Gini coefficient -- it's a digital format which is confusing to use, easily delegated to centralized exchanges and, most importantly, reliant on multiple layers of software and hardware trust. What if instead Bitcoin and Ethereum were like cash, a physical format that billions of people are intimately familiar with? To achieve a viable physical cryptocurrency, however, we need radically open hardware with cryptographically-secured guarantees of their face value and uniqueness anchored in silicon. We propose a novel, low cost, physical format for cryptocurrency -- more appropriately "cryptocash" -- that's intuitive to use and allows for free, anonymous, and instantaneously final transactions. By isolating cryptographic key material and operations from software and firmware in secure silicon, we can build a new hardware foundation that supports not only self-ownership of crypto assets, but broadly serves to secure the secrets users need for participation in the decentralized web. We'll hone in how shared trust in hardware and chips must ultimately be quantified through crypto-economic incentives such that we can move away from manufacturers that willfully or ineptly include backdoors in their products.

Speaker(s): Cameron Robertson
Skill level: Beginner
Track: Society and Systems
Keywords: storage, wallets, general

Follow us: https://twitter.com/efdevcon, https://twitter.com/ethereum
Learn more about devcon: https://www.devcon.org/
Learn more about ethereum: https://ethereum.org/ 

Devcon is the Ethereum conference for developers, researchers, thinkers, and makers. 
Devcon 5 was held in Osaka, Japan on Oct 8 - 11, 2019.
Devcon is organized and presented by the Ethereum Foundation, with the support of our sponsors. To find out more, please visit https://ethereum.foundation/

## Transcript

I'm Kevin Robertson and I'm here to talk to you today about how crypto is broken one of the big things that people talk about in terms of adoption of cryptocurrency are things like transaction speed and transaction volume and cost but we actually believe that's not the core reason why people aren't using crypto broadly today we believe that using crypto for most people is dangerous if you look at a lot of wallet apps out there all the warned users about their private keys you set up a new wallet store your private keys safely if you lose them you lose your money now for technical folks that might be okay but for most people that's simply terrifying the result of this is that only the most technically savvy users know how to use crypto in a self sovereign way exchanges are the default for the rest of us people go to exchange as a bi crypto and we all understand the risks that are associated with that hundreds or tens of billions of dollars that have been lost on crypto exchanges over the past seven to eight years so how is this shaped crypto and what we actually do with crypto today well the simplest example is people don't use crypto to buy real things instead it's there for speculative behavior along with that crypto has this obsession with TPS we need to focus on transactions per Section we're per second to get more users sort of if you build it they will come mentality but our thought is we're getting close to the chasm of crypto and in order to cross that chasm safely if we want people to use their crypto on their own and hold on their own we need a new solution so we asked the question what if crypto was more like cash cash has a lot of benefits to it billions of people understand how to use cash transactions are instantly final they're largely anonymous and if you destroy it or lose it you gone it's gone so if you burn your your if you burn your cash you no longer can spend it along with that cash is still massively used so a lot of folks talk about the move to credit cards or spending things digitally but in most societies cash is still used for transactions under $25 in order to think about if we want to make crypto more like cash we have to understand what cash is and what it represents it's something distinct from money cash is a transferable IOU if I hand this cash to you you know that you can go hand it to someone else in exchange for good or service it means that you have to trust that cash is real and that someone else work will accept it it's also transferable there's been a number of experiments in this the easiest one is printing out a private key and handing it to someone that's great it's incredibly cheap but it's also technically difficult to use and beyond the first person it's not infinitely transferable to more and more people other people can simply copy the private keys the other solution is something like a hardware wallet hardware wallets store the crypto in a fairly safe way but the problem is they're not easily transferable and they're fairly expensive and along with that you have to trust the stack of the hardware wallet you have to trust its firmware and its software recently there's been a move towards Java smart cards Java smart cards are great because unlike harbor wallets they're fairly cheap but you also can't trust them either they have a full Java smart card operating system running on them with massive vulnerabilities what this really boils down to is that trust in usable crypto comes down to trust and keys and key storage and so what we're proposing is something that we're calling call calling is an experiment in physical crypto cash it's transferable it's trustworthy and it's low cost maybe physical cash is what we need to actually achieve price stability out in the world against crypto rather than stabilizing crypto against fiat which is inflationary we can have crypto which you can actually use to spend on goods and services let's talk a little bit about how Kong works it has a printed face value on it it's really simple to understand how much you have and how much you can spend it's tied to a smart contract so unlike a hardware wallet it's not storing the asset directly on it but rather the smart contract does that for us and what that means is that we get all the security benefits of the EVM it's got a time lock on it crypto cash is not a prepaid gift card you can't pull the crypto off of it immediately that's a really important feature because I hand this to someone else they have to have a good expectation that the crypto is actually backing it and lastly and most importantly a secure element chip and we'll talk exactly about how that functions in a moment this is all somewhat familiar because in some countries cash used to work this way in fact in the early 20th century you could actually take certain notes and exchange them for precious metals at the federal reserve crypto cache is similar at the end of time lock you can take your Kong and you can retrieve yourself the token that is backing it so let's come back to the secure element the secure element is the most important part of crypto cache because it's where it stores the keys but there's a few issues with secure elements today many of them allow you to load code onto them but there's a handful that don't they're only configurable you can generate an ECDSA key pair get the public portion but not the private portion by design the private portion is locked up forever you can get signatures out of it however to verify that in fact that device is the authentic device containing those keys the only problem with this is that existing secure elements that operate like this run on P 256 P 256 is not the native curve used in Bitcoin in theorem K 1 is but the great thing is we can use a smart contract to bridge the two so what we do is the smart contract is the custodian of the tokens for the time lock period and then it also has the public key in P 256 form because of smart contracting we can have a P 256 proxy con tract which understands that key and can verify the signatures for us broadly we believe that Kong is the first of a new kind of device that we're calling silo effectively there's silicon locked contracts they remove all of the security required in entrusting a hardware wallet stack and they put it on chain into the EVM if you can trust in your chip and your key storage everything else is open and auditable you don't need to worry about some firmware bug you don't need to worry about someone messing with your hardware wallet in transit this could be used for all sorts of applications not just money you could do this for asset tracking you could do this for art NFTE etc etc but this brings us to the next conclusion which is if this is the right way to store crypto in a usable way that you can hand from person to person we need to think about the chips that we're putting it on to and the next step is that we need to look really closely at opening up that silicon this is an example of a chip that was in a Samsung Galaxy product from 2011 and here the designer took some liberties and actually putting art at a gate level into the chip now that's all nice and fun but what if instead he put in a few gates that extract the private keys millions of people might be using this device with really secure open-source stacked on top of it but if somebody has ability to pull the private keys off of it all those guarantees are for naught we're proposing a new project which we're calling arks and I don't have enough time today to go super into detail on arcs but effectively the notion is this we need to think about trustworthiness and our chips and we need to put bounties behind them so think about it this way if you have a chip out in the wild and it says it can only store private keys on it then that's what it should do and if someone can extract the private key then the bounty is yielded we could use this model for much higher things in the stack as well but if we can do it on silicon chips we can create incentives for people to start breaking them and saying if I can make money from this that's useful securing crypto is only the first step though one of the big pieces of this is that Keys underpin everything so when you log into Google or Facebook or even a decentralized service like using PGP or signal or tox all of these things are relying on key storage and so if we can have better key storage we can change the dynamic so that when you go and setup a user who's not technically savvy they can have key storage which is tangible and understandable and way less risky than a private key that can be extracted or pulled off there's something really interesting that's happened since 2009 in 2009 we had this breakdown and Trust in the financial system which arguably is one of the big reasons why Bitcoin took off in the ten years since then however we've had this rise of all of these online apps and online services in the web 2.0 era but we're finally seeing in 2019 is people are realizing that maybe that wasn't such a great idea maybe trust in Google may be trusting Facebook with all this data wasn't wasn't what we needed when Snowden revealed that that even if we use things like HTTPS when the government inserts themselves in between servers we can't trust that our data is not being spied on now everyone in this room knows about this and we're all working to solve this problem but if people can't store their keys in a way that's understandable then how how are they not just going to go again and delegate it to a third party in the same way that they're using exchanges this brings us to a few simple conclusions the first is that we need usable self sovereign crypto people storing crypto on exchanges is not the way that we intended for it to work if people continue to do that where we're just gonna have more breaches and more losses and decrease the overall trust in the system and moreover exchanges are really there just to encourage speculative behavior but if instead we want people to spend crypto having in a physical form that's easy to do that and trustworthy is really important the second is that to achieve this we need an open hardware stack starting at a silicon level we need to look really closely at how our chips are made and we need to think about incentives behind those so that if I go out and build an open chip and I I discover after the fact that that chip has a bug in it or someone's built a backdoor into it then that there are crypto economic repercussions to that somebody effectively has to pay so what this brings us to is that what we've designed today komm it's physical crypto cache it works and it's the first silicon locked crypto but this is really just a demonstration of what we can start to build for the space in terms of all sorts of silicon lock crypto for things ranging 2nf teas to crypto art to then all of the keys backing our services so that our users don't delegate them by accident to other services which in turn will break break down the trust thank you very much for your time today come find us we have the metal briefcases and we're handing out kong and we love for everyone to test it I'm Cameron Robertson thanks a lot [Applause] you