Privacy Highlight: We Almost Didn't Get Encryption

all right I'm super excited to introduce you guys to our first panel it's called we almost didn't get encryption a lot of folks have been speaking in the past year about how we may be entering a kind of second crypto War uh that may have kicked off with some of the you know events over the past year I think it's important to remember that there was a crypto war in the 90s and we have some of the survivors of that war here some true died in the wool cypherpunks I'm super excited to learn from them about what happened what you guys got right what went wrong in the 90s ml uh is going to moderate the panel take it away thank you testing thanks for the overview Lane um we take encryption for granted in our everyday lives and I think we're so lucky to be able to do that and everyone sitting here is familiar with asymmetric key cryptography we're familiar with public addresses and private Keys that's it's really shocking for something so mathematical to be so broadly at least at a high level understood but I was shocked I'm younger than most of these guys here so I was shocked to learn that this form of math was deliberately intentionally proactively kept out of the hands of the public for so long so today um I want to take a trip down memory lane and talk about how we got to now how we got to you know be at this point where where we won a few battles in the crypto Wars we're certainly not done and two of the people on stage here were actively involved as part of the group that helped us win some of these battles The Cypher punks I'll uh I'll let them uh introduce themselves individually and I did ask them to prepare a personal anecdote related to encryption Ryan you want to go first hi I'm Ryan Lackey I was involved in the cyberpunks from pretty much a couple months after it started in the early 1990s I got involved in crypto because as a young kid I had no power I discovered cryptography where individuals have more power than everyone else in the world for all time put together and it was really attractive to me my interesting uh anecdote about this is I emailed uh on the list there was a usernet group called psychrip that had information we could order from NSA all of their security documents uh the all the open source documents so as like a 13 year old kid I called NSA on the telephone and ordered uh like three linear feet of the rainbow books of security guides from them and then used that 10 years later to build an offshore data Haven Anonymous electronic cash all sorts of other crazy stuff I'm so not following that story that's awesome I'm Dean Tribble uh I was relatively young at the time of cypherpunk starting up and um and have been working in computer security for quite some time and then came back to to blockchain to apply our How We Do programming but but when I was living through that this was you know I started pre-internet and privacy was normal and so when when these things were coming out and we're sort of hearing this technology like well but there's this awesome new stuff that Ralph you know was involved in before he was allowed to talk about it Ralph Merkel right and and we're trying to play this and what do you mean you have to do 40-bit encryption I just don't even I mean why would you do that to yourself and yet there were all these export regs and it was a munition and and and suddenly if you were trying to to build into your business system the ability to send transactions um you couldn't uh or rather you would end up using this stuff that that you could break on a normal computer even at the time in a couple of days and so we there was a system that we had for doing distributed programs and you couldn't use encryption so we had or you could only use four-bit encryption so we had the Daffy system which you would be silly to use that was produced in the US and there was random people outside in Angola um that they would take it and replace the encryption that we could not use or the 40-bit crap with real encryption that of course was available outside the US but you couldn't have it here and we just had all of these turn you know these crazy rounds of trying to navigate that that stuff that I was you know Tangled Up In so hey I'm David Campbell most people call me DC I've been fortunate enough to found a number a handful of startup companies all in the security and privacy domain I was just a kid during the crypto Wars of the 80s and 90s so I looked on with curiosity but mostly as a bystander but I was pleased many years later to be a c-level executive at a web scale infrastructure company where I had the opportunity to actually encrypt a small but meaningful part of the internet this in response to the revelations of widespread Dragnet surveillance being perpetrated by various government agencies I believe that privacy is necessary for society to progress but I also believe that we need to be thoughtful and pragmatic about the path we choose and the trade-offs we make as we build this next chapter of the internet thank you hi my name is Sherry Davidoff I'm the CEO of lmg security and I started off over 20 years ago on mit's network security team which was where I first started analyzing traffic of course to respond to different incidents and my undergraduate project there was to analyze all of the University's Network traffic and I think when you do that on a day-to-day basis and you see just how easy it is to dip a toe into that stream of data you then understand the importance of privacy the importance of putting your emails in envelopes and so when I started my company lmg in 2009 the zeroth thing that I did not the first thing the zero if because it was even before the company started was to make sure that we had full end-to-end encryption in place automatically no one could mess it up for all of our internal emails using pgp and it took a while to put it to piece it together to get the right mail client to make sure everybody had the right plugins but I'm sure over time that this would evolve that it would get easier and it hasn't it's actually gotten harder and to this day we still have full end-to-end encryption in our email system I don't know of any other company that does that at least our size but I'd like to I think we need to work together to make it easier and to make it more accessible for our own security and for National Security and the security of our globe thank you that's great so many Amazing Stories thank you all um so I think it's our first topic of discussion if we can rewind ourselves to the 1990s maybe Dean you can start off with kind of a short overview of what what the cyberpunks were trying to achieve yeah as uh you know previous speaker said privacy is about having the choice of what you disclose and so much of cypherpunks you know there there were whatever you know Tim May was the shock jock of cypherpunk where he was you know he would rant about crypto Anarchy or what have you but but most of it was you know we have the technological ability to improve our world to improve our privacy and so the Mantra of cypherpunk's build code well you could build software or you could you know build a plan for for for for political Improvement in the scenario but The Cypher bug meetings people would get together and they would share what had they done right you know there's there's articles where you know people would talk about anonymity but as we see in the in web 3 often it's pseudonymity and so I had you know there was lots of the early discussions about reputation systems because of someone pseudonymous you don't know who the true name is like the the Werner Vin story you don't know exactly who it is but you can know it's the same person from last month that did you you know the did you assault order that's been building a business over time and so you know so um from anonymity to pseudonymity and reputations and complex systems of how humans can cooperate we can build systems so they can do that and stay private and that was a big chunk of what we wanted to do in cyberpunk that's great so when I think of cyberpunks I think the well-known story is you know they they worked on cryptography a lesser known at least in other circles is our remailer efforts and I think Ryan over here has really interesting stories about that sure yeah one of the so the interesting thing about The Cypher punks list is pretty much all the things that are getting built today were discussed on this mailing list in the 1990s in Prototype form there were a couple of enabling technologies that didn't exist at scale uh good quality handheld computers that were secure that you could easily build software for uh easy ability to write software there are people like Microsoft that were gatekeeping writing software on the desktop at the time and easy server platforms all of those have been solved so we're now seeing an explosion of stuff retailers are one of these things another one was the anonymous electronic cache which we now have sort of in various ways coming I mean bitcoin's anonymity is is a definitely a an issue and all the the global Ledger systems but but that was an enabling thing but yes remailers were one of the the most interesting things that actually wasn't purely theoretical at the time you could actually take software run it on your computer and store and forward message images in a way that was very very difficult to trace I ran a remailer on my server that was a PC running on my desk at MIT I ran one offshore in the offshore data Haven I built later and I got requests from law enforcement around the world all the time for for tracing messages and I got to give them the best answer possible which is I don't have the ability to help you not just I don't want to help you and there was a case later I did some defense Contracting in Iraq and one of the FBI guys who was on the same base as me knew one of the people that was calling me about the remailer was still running and we went out and we we got a drink it was a non-alcoholic drink on base and I just told him like yeah this is how it works is you don't get anything and they stopped asking questions which was really nice so there's these these systems if you build them well and you build the technology correctly you're not going to have to be brave yourself I'd rather be brave up front about building the right software rather than being brave when there's a potential legal process facing me for how I operate something and remailers definitely got that right they were great because they had immediate value they were difficult to build but they were one of the early successes and hopefully there's a lot more so thank you for that because I worked with Eric Hughes on early remailer stuff put one up but I was a student at the time so then I moved and my bandwidth stuck so I just couldn't you know I couldn't I I wanted one to use and and couldn't run my own for a while so that's great I'm I actually have a piece of the cypherpunks manifesto from 1993 in front of me I'll just read the first piece so so you give context we The Cypher punks are dedicated to building Anonymous systems we are defending our privacy with cryptography with Anonymous mail forwarding systems with digital signature signatures and with electronic money so it's a shout out to all of you of you you're still part of this um okay thanks so much for that and so once encryption was out of the bag right the the government then thought okay it's out but then how do we insert a back door into that can we move forward from the early 90s to Mid to understand what happened there with the Clipper chip for example if DC you can take the lead sure so Clipper was an attempt that I believe was well intentioned but poorly executed so the the government the NSA specifically was observing that crypto was becoming ubiquitous and the Signal's intelligence that they had been using for years and years was become was at risk of becoming ineffective and so the idea was well with Dez they had artificially constrained the strength of the key the Dez encryption algorithm was originally designed by its designer with a 128-bit key the NSA as part of that approval process said no we're going to have you ship this with a 56-bit key which they could certainly crack in a reasonable amount of time but things move forward a t was shipping a phone that was going to be strongly encrypted using public key encryption in strong algorithms and NSA really wanted to get out in front of this and thus was uh the Clipper project was was started and the idea here was key escrow so we're going to have a strong algorithm that nobody can break but the authorities are going to have a copy of the keys required to decrypt information that is approved by a court order to be surveilled now the problem with this is multi-dimensional but it feels to normals a lot like the government saying okay you can have locks on your doors you can have privacy in your home but you're going to be required to keep a copy of your key at the local sheriff's department so this this obviously didn't feel good The Cypher punks pushed back uh the Big Brother inside stickers proliferated but it wasn't until a researcher by the name of Matt Blaze hammered on the uh the Clipper implementation he actually found a flaw not in the design but in the implementation of the surveillance mechanism the escrow mechanism that allowed the checksum to be uh basically collided and that blew up Bill Clinton having recently been elected was supporting this because he thought it would help mitigate National Security threats but in the face of the sort of quick hack that Matt Blaze was able to put together the widespread privacy concerns the Big Brother Dynamic of all of it was really unpalatable so Clipper was shelved uh strong encryption became ubiquitous and I think we have now seen the industry have to move they can't just surveil all the traffic they have to do detective work you see this now with the Silk Road you see it with the alpha Bay guy they literally sit there and watch they they profile their habits because they know they need to get that laptop with logged in and open if they're going to have a case so uh I think that's Clipper in a nutshell in the case of the Clipper chip it was about back door and then when that failed then they went on to try and weaken standards with des and then AES we don't have time to get into that today but um Sherry maybe you can talk about you know based on the work you do we get cultural standards you know backdoors what can you tell us about that well when DC was talking I was just thinking about the fact that what came out of this it's not that the concept behind the Clipper chip went away the Clipper chip itself died but we unwittingly gave law enforcement and intelligence agencies incentive to do things like stockpile vulnerabilities and Target the end points you look at cases like the shadow Brokers and leaking of uh of vulnerabilities like Eternal blue they're keeping those to themselves so that they can get that backdoor access and Target and they're looking for things like keys they want to steal keys they want to Target the endpoint and yes we've also seen some deliberate weakenings of the protocols themselves but all of this impacts our security in so so many ways for example I do a lot of work with Healthcare I always have I worked at the Children's Hospital Boston and we all care about our medical privacy right and when HIPAA started to really get enforced around 2010 2011 all of a sudden sudden you started to see a huge uptake in encrypted email systems right and are these really it drove me nuts because my clients would say oh man I hate encrypted email it's so hard to use and I was like that's not email you're getting a link in an email you're clicking on the email you're going to a website people are securing the website with a password like fluffy one or something like that and it's definitely not more secure and it's definitely not email it's a pain in the neck for everybody and you could see you know I started wondering it's easy for me to use encrypted email on a day-to-day basis once you get it set up but everyone kept saying oh it's so hard it could have been easy right guys Microsoft could have chosen to put a button in the system that said get public key or just get it automatically right but those choices were not made and there's a number of reasons to do it maybe we can go into it more but you can see that that weakening of standards and the ineffective deployment of encryption throughout our society has really bitten Us in private sector security as well that's great we have about four minutes left I think it's a great time to tie that to today you said Microsoft could have made it easy to click a button and have everything be end to end we're we're seeing today's crypto Wars right with in 2001 very recently when Apple wanted to scan iMessages and and all the photos you have on iCloud and then the UK is drafting an online safety bill which basically requires um any service provided people will provide in I think plain text you know or the underlying messages and the they're saying it's for CSM and other kind of law enforcement reasons can we talk about what you see as kind of today's Wars let's situate these people back into the fact that it we we urgently need to stand up as as individuals as companies first I'd like to hit a meta point on some of these things in like the Clipper chip or in you know there were papers that that talked about how key escrow as an architecture doesn't work there were people in the government stepping up to say this system is secure or we will protect your data or what have you and that's one of those things where that's one of the first red flags to look for is where other people are claiming Authority they should be providing you real information and they've got strong interest on another side or they're going to create a database that they claim will be used according to one policy but something happens and and it's politically expedient to use it for something else and and then they will and so we saw that then we're seeing that sort of thing creep in now it's exactly the give your master key to the to the sheriff's department and so all of these systems as we start heading there could be a problem in the case of Microsoft because I was they actually acquired a company where we were doing encryption for businesses trying to do as end to end as you could but it's most but it's for businesses where that could create liability and so what you wanted is to have more access inside the business in a reasonable fashion be interesting to talk about the details that but figuring out the use case is hard right figuring out all the cases of wait you just signed something but I didn't mean to sign that as a business or we need to recover it and you're sick and I can't get your keys or those kinds of things it there are some challenges in all the more interesting ways that we want to do encryption but man individual encryption that ought to be easy and it's not yet so thank you I think the biggest issue with getting crypto adopted has been user demand like back in the 90s users didn't really care they didn't know about that they should care so they didn't care they had lots of other things to worry about and even today computer systems are hard enough to use as an individual as an organization everything else that security and privacy doesn't get prioritized I think we owe a lot to our our friends at NSA and other organizations and what happened in with the Snowden leaks as well as all the other Revelations of yes they are actually monitoring you it's not a hypothetical thing and as well as all the the hackers ransomware projects everything else that are that are causing people to actually care about security now there's a real market demand for it that's going to fix the problems just a question of whether this is going to be two years to fix it five years ten years or 20 years but I'm confident now that there's clear demand for this it's going to get a lot better so we've got right now stronger encryption than we've ever had but I think we're in a worse position for privacy than we've ever been and so for all of you here that are building I implore you to think hard about the data that you're storing don't collect it if you don't need it don't keep it for longer than you have to to serve your users and think about privacy as something important for the the evolution of society if we do this right we can navigate web3 in a way that will be Pro privacy and we're just not heading in that direction right now so we need your help data is hazardous material and of course when it leaks it can cause a lot of damage in many different ways and I just wanted to encourage everyone who's building things to think about building in end-to-end encryption and allowing people to have their own control over their data over their money over anything else that you're storing we've really when when Snowden when the issues with Snowden came out we saw an interest in privacy all of a sudden Google came out with an end-to-end encryption plug-in which I thought would be a huge advancement and then it kind of quietly died and it was exciting over the past year to see that re-emerge again I think there's a lot of interest in privacy now there's more awareness about the need for end-to-end encryption so just something to keep in mind because it's a powerful tool and it can help protect all of us all right that's all the time we have thank you all so much