Episode 306 - Predictions: ZK in 2024

Transcript

[Music] welcome to zero knowledge I'm your host Anna Rose in this podcast we will be exploring the latest in zero knowledge research and the decentralized web as well as new paradigms that promise to change the way we interact and transact online this week I get together with my amazing ZK podcast co-hosts tun garmo Nico and Kobe to do a look forward at 2024 in ZK we cover emerging ideas for the coming year research predictions dream applications the best case

scenario for ZK the worst case scenario for ZK the most likely case black swans and more we also do a very brief look back at 2023 with Nico and Kobe who are missing from our last episode and we give you some insights into upcoming events speaking of upcoming events before we kick off I want to share a little bit about ZK hack 4 online this 4-week event runs from January 16th to February 6th this is not a hackathon but more like a crash course in ZK it's pretty unique as an event as it combines three things over the span of four weeks it is a live virtual Workshop Series where we meet every Tuesday to learn about new ZK tools it's also a CTF like

puzzle hacking competition with three puzzles running over the 4 weeks and it's a job fair where you could connect with and get to know your future employer in a very casual virtual setting in this episode we're going to be sharing more about the puzzles themselves and how they're going to be running so stay tuned to the end of the episode where we cover that the geometry research team has built the puzzles this time around and so they're going to be sharing a bit more about what they have planned we are very excited to announce our Workshop Partners risk zero who's presenting on January 23rd and polygon Labs who are presenting on January 30th be sure to check those out and also join us for our first session happening on January 16th in this session we go over

the entire event how it runs what to expect we will then be doing something called the ZK jargon decoder session and this will be me Nico and Kobe you'll hear a little bit more about this as well later on in the episode and then at the end of that first session on January 16th we kick off our first puzzle so I really recommend you join us for that just to get to know this event this is our fourth time running this online event it's a great way to get to know the ZK hack Community from all over the world without needing to travel and yes I hope to see you there I've added the link in the show notes now Tanya will share a little bit about this week's sponsor Alo is a new layer one blockchain that achieves the programmability of ethereum the privacy

of zcash and the scalability of a rollup driven by a mission for a truly secure internet Alo has interwoven zero knowledge proofs into every every facet of their stack resulting in a vertically integrated layer one blockchain that's unparalleled in its approach Alo is ZK by Design dive into their programming language Leo and see what permissionless development looks like offering boundless opportunities for developers and innovators to build ZK apps as Alo is gearing up for their main net launch in q1 this is an invitation to be part of a transformational ZK Journey dive deeper and discover more about Alo at al.org and now here's our epis [Music] episode today we are doing our look

forward at 2024 and I want to welcome my amazing co-hosts we have Nico Kobe tun and GMA with us today welcome everybody thank you thank you happy to be here what's up Happy 2024 oh yeah yeah we're here so the plan for today is to definitely look forward at 2024 in the ZK space but we have two of our co-hosts Nico and Kobe who were not present in our last session where we did a look back at 2023 so I actually want us to spend some time at the beginning of this episode talking to them about what they experienced or what what sort of stood out to them in 2023 and before we do any of this any look back or look forward I

usually don't have to mention this because we always talk about tech but given that we're doing predictions I will just say a quick this is not advice or anything in what you should invest obviously I feel like that should go without saying but I figured I'd throw it out there since we are kind of trying to predict yeah okay so let's kick off our look back at 2023 Nico Kobe what are you thinking yeah well for me research- wise 2023 was super interesting um we had the whole folding Saga at the beginning then we had all the like the return of gkr you know like all this like lasso stuff log up stuff um and then around the end we had Bas come up

so I think 2023 was super research heavy and that's part of what I'm looking forward to in 2024 it's going to be like people experimenting with all this stuff what about you Kobe what do you remember from 2023 yeah I think I think that um the the last year I gave it a theme and I'm trying to remember what it was was it the year of lookups or was it the year 2023 no 2022 22 I think you would have said lookups yeah yeah yeah and I think that um that recently what what we've seen is that a lot of new falling work has come out and I think that there is much bigger hope that we can use

proving systems for use cases that haven't been possible before and yeah we'll talk about it more later but more generally I've been very encouraged to to see topics like identity come up uh for 2023 that was very very strong for me mhm and something that I've been waiting for for many years and additionally the ability to use ZK by people that not have ZK training let's say it like that and I think that's also something that has become just more recently practical so those those have been really influential for me cool would you say though that 23 research-wise was the year of folding

good question I need to to review it but I I like I like making these themes but let let's tentatively call it like that I guess okay except for at the end Bas comes in which something else and then he shuffs all the cards right so that's the fun part we can give it two halves right yeah yeah first half is folding and second half is generic ECC based constructions airor correcting code not i' go even further like for the past years we've been obsessed with putting proofs on chain and having smart contracts verify proofs so in that context you want like the smallest possible proof the shortest possible verifier and we sort of obsessed over that um so you get to gr 16 and we're happy but if you start dialing it back

and be like oh I want proofs happening on I don't know a consumer laptop and I don't care who verifies then you dial it back and you go back to things like gkr Bas or like back not really back in the case of Bas but you start loosening the requirements of small proofs for the requirements of fast provs and I think that was the theme for at least the second half of 2023 even folding actually is sort of in that direction mhm mhm I want to also bring up something because when we did our look back at 2023 last month I hadn't really prepped that much for that like we we just sort of spontaneously were together and we're

like let's make it the 2020 23 look back but today I actually had a chance to listen to our 2023 predictions extensive preparation yeah and I want to bring up like two three things that we covered so tun your big question at the time was like what is the what is the asset class kind of the asset that will Define the next year and to give examples of previous assets You' been like um sort of the lp token the the strategy bearing tokens or or asset like things like what urine was doing those were past ones and yeah you were trying to predict a future one you had one about ZK uh oh which was

about because there are none of them now I don't think it happened but it was something like being able to go cross chain with ZK proving like liquidity on one chain to another do you remember this at all uh yeah I I I I vaguely remember saying that but I I don't remember precisely what I for being totally honest totally putting you guys on the I I I will say I will say I think that like sort of happened I feel like some of the bridges do kind of do that but they all are kind of still half asked like I don't think they've crossed the chasm to what we talked about so yeah they're not like the defining asset G you had one which was future claims for interchain me ah cross chain PR also

doesn't no one cares about that nobody cares really oh [ __ ] it's just there's not that there's not that much of of cross train M that's right not yet yeah not yet maybe at some point well I I I think you need the former thing of like the bridged liquidity first then people are fine doing it but I feel like no one wants to take the risk of waiting on the on the bridge right now was there some big theme for 2023 that you could kind of summarize in that context in that thing you had sort of suggested TR something that did work what is the asset of 2023 all layer ones so I don't like there's not really much unfortunately all the assets that that

got liquidity were not necessarily new assets they were the old assets getting new money okay it's sad sad sad 2023 well cuz like cuz like cuz like I feel like a lot of the tech tech technical stuff didn't lead to new assets this year like there was a lot of technical advancements that like didn't you know instead we got a ton of dog [ __ ] meme coins right like we didn't get we didn't get like with much we yeah exactly we got like dogs with Hats right we didn't get like we didn't get like ZK asset we didn't get these kind of technical but this podcast is about the high futin technical true true desires it sounds

very like late Stage bar mark Market that you were this sort of like rehashing people were exhausted that's why it's so exciting that we're now in 2024 we can properly put I mean 2023 wasn't bad 2022 was rough 2023 was okay but I think it's still great to put it behind us and now we can start looking forward I had a few questions or I I basically sent you all a few topics that we could cover what I want us to do is a prediction and obviously predictions are flawed and we're going to be creative here we don't know what's going to come out correct what's just an idea maybe something we'll predict Now isn't really a 2024

prediction but a later prediction we'll see what comes the first one that I wanted to talk about were the emerging ideas that we think could come in 2024 and so these are like yeah it's a it's a pretty broad like what is an idea it's like concept this is like Theory proposals that we think might come this year I mean I have I have one that I it's like more of a deser DA rather than like a real thing but the merging of what feels like two very different currently strains of ZK mhm uh into like one in a very deep way so so specifically something like like beus or like fry or like you know any one of these uh with some kind of highlevel

like fancy cryptographic algebraic structure right like the kind of currently they're very separate things there's like things that use deep cryptography there's things that use uh deep results and AER correcting codes and those are kind of different uh for the most part not not quite 100% but for the most part different um I I think we're I hope maybe it's better the way of the better way of saying it is sometime this year we'll see like a deep Union between the two uh in a way that's actually you know it's not what people are currently doing which is they're proving something thing for example in Fry and then using a second prover to construct a shorter proof that goes on chain via some kind of deep cryptographic techniques so that's that's my my prediction my like

research hope for this year is to see like a really really deep protocol that uses both things uh and structure from both things in ways that are like very not obvious currently nice one one thing I've been thinking about and I think it's ever since we did that episode Kobe with um Costas and aush mhm um this idea of more Web Two into web 3 more porting of identities from one to another I know I mean years ago Mina was touting that with ZK apps what became ZK apps was this idea that you could all of a sudden prove things in web too but I felt like yeah with that experiment I just it seemed to me to be opening up something towards this this year where maybe

instead of just talking about like the technical of how to do that we actually get to see use cases where it's super super useful and interesting and yeah I I had that down too completely yeah yeah ZK ZK login um and all the sort of proving TLS session stuff so like TLS notary and all the different constructions that do the same thing um I think those are going to be super important similarly I think the identity work is like super important I'm I'm not sure how much on the research side yet I I do see that there's going to be advancements in let's say the implementation and performance side and maybe some breakthroughs as to how to connect it it in a more holistic way um but some

things that maybe it's less of a prediction but more of a hope um is that some cryptographic Primitives that have been less how to call it less profitable uh or like non non snarks so unrealized profit unrealized something like that so like verifiable NPC and potentially verifiable FHA um those might end up being important for use cases later on and I'm seeing seeds of interest in them like concrete interest and maybe there's going to be either a breakthrough or at least some order of magnitude Improvement there would you say the

reason we haven't seen that pick up is maybe because you can't make a blockchain with it uh yeah I think that's definitely one of the things or maybe you can they just haven't figured out how to yet uh I think some people are thinking about it but uh yeah it's not it's not as obvious uh it's interactive sometimes and it requires some new communication model that is not as convenient and I think more more concretely one of the big explosions that we've seen for ZK was scaling and there isn't something that is super figured out for those Technologies yet and now we see that people are having ideas as to what to do with these things

and they see that the community around them or the public starts to care more about the properties that these give you will potentially move away from something centralized eventually uh not not immediately but eventually so I think the combination of this might might lead to to more interest there we'll see I I I have the opposite take unfortunately despite despite wanting to that to be true for like for personal reasons uh which is I I actually think we're going to see way more te stuff this year unfortunately and like O oam type of things I think we're not going

to see as much as of other stuff like like in practice I feel like there definitely feel definitely feels like I've I haven't seen that as much interest in Resurgence and interest in Te stuff as I saw in like 2018 I feel like te's were like kind of dying maybe there was a little bit of a Bo bump with flash Bots but then like it seems like a bunch of people are like ah we're just going to go back to te so my counter to to Kobes is I actually think the unfortunate like aam's razor aspect of like well we'll trust Intel uh is is is going to happen you know until the next paper that says hey sgx is completely broken and then the Hope dwindle again

then the a week you know I don't it'll be it'll be for exactly I feel like the halflife of that thing is decreasing as as the more exploit it's like it should be increasing with the number of exploits but it's decreasing which is not good by by the way I wouldn't be sad if people do T is but they do it in ways that are better than they're doing it now and they have better security models and they threshold eyes and all of this I think that would actually not make me that sad if if people use these combine I mean it's a defense in depth kind of mechanism combining it with other techniques exactly on another topic for 2024 emerging prediction here is like I think it's already kind of happening

where all the rollups are ZK rollups eventually what do you mean by ZK some form of ZK like they're going to incorporate ZK yeah somehow I mean I think we've seen it with some of the optimistic big optimistic projects like investing in ZK I mean I know it partly because they'll reach out to me all of a sudden for like the the summit or something where I'm like you guys aren't ZK you know maybe they are a little bit so yeah I wonder what do you think well how fast or real do you think those efforts are cuz I actually think there is there's like many things moving at the same time it's not I'm not convinced they're like pot committed let's put that way I that might be true yeah maybe they're just dabbling just exploring I

mean how committed are they to fr proofs in general that's also a question to right I I I think there's a lot of steps in along the way here that make it very the make it a little more I'm not 100% sure it's worth I I wouldn't quite stand by the everyone's ZK rollup statement I guess my question is like why did it become a thing now you know why did everyone decide to because it does it does feel like it was a it was a wave right it just kind of a bunch of people were like yeah we should just do the syn proof CK thing because bisection are impossible to audit like honestly they're really annoying to audit like have you have you looked at the the some of the bisection code base it's like I want to kill myself read what is that is

that a fraud proof is fraud proof B it's it's like the sort of Base algorithm within these types of FRS and okay it's it's just like it's very hard to reason about the the changes to State especially to particular applications in in a way that I I feel like it's it's yeah the security being 100% sure of the security of it's really hard but then again like hey if if no one really cares at all about these things maybe but one of the reason could also be like the the ZK rollups War are over I guess the first ZK rollup is out and now these people who are like oh I'm not going to get involved suddenly think like okay here's the space I can get into aren't the snarks also just like

better like how to use them the sort of like down the line consequences have been mapped out a little bit and so a lot of these teams are like oh that's how it works yeah I mean if I say Snorks are bad I go out of business so I kind of have to say that they're great they're much better they're the best well well I I I think another way of thinking about it is like suppose you did optimistic you got a huge market cap you have a ton of usage the DK rollups look like dog [ __ ] comparison like market cap and usage wise for for and for reasons not that are not their fault they're later to the game they like it's hard to convince people to move their liquidity uh incentive programs aren't as sticky are stickier in in in certain

Assets in certain ways Bridge liquidity is more annoying like there's not as much you know there's tons of reasons that they they're not like fundamental reasons like you know you could imagine the universe slightly mutating the time of timeline of everyone and then you know we would be talking about scroll or or something like that being like the largest tvl thing but you know whatever that that's just the way it is but once you have a large market cap large tvl it becomes a lot easier to be like hey I can add all these features other people have right and I feel like that's really more what's happening but it's not clear to me that there's like commitment of like hey this is yeah yeah it's more like it would be cool if we could do it

yeah do you think there could be some other thing that comes up this year that like replaces both of those things is there anything replaces or augments we're talking economic versus cryptographic but is there another kind of cryptography I mean yeah you could you could do both right you could do like whatever uh proofs or something n proof yeah yeah yeah so it's a it's a mix of both uh although it's quite a bit of overhead but it' be it'd be kind of a fun thing I do think like the problem with a lot of the we do both things is the wall clock implementation time of that just like I I mean like I said just like reading some of the optimistic fraud proof code I actually feel like it's conceptually

harder to follow like the the state machine of what's Happening than a ZK VM which is like crazy to me yeah I I feel like you should be able to make like a very nice abstraction for like rolling back a blockchain because kind of you have to do that all the time like whenever e for or whatever but the problem is these like bisection proofs are sort of like not as simple as you think because I have to decide whether I'm invalidating state or execution or you know like and like trapes through both of those in both dimensions it's and like in the ZK case you kind of like okay I can assume this checkpoint is correct from here onwards in the bisection case there's a lot of like oh like do I need to reload the state variable or not and like is it valid at

this time it it it's it's quite a bit more hairy because you can't trust the you you either do the N squared replay which takes forever or you do this kind of like I keep track of like what I've touched and dirty then not dirty then it's it's it's quite a quite an annoying I I I just think it's like conceptually simple but oh just go look at the unit test for these things they're they're they're hairy there another thing to remember possibly about doing both which is that you might gain a lot by doing the fraud proof part in ZK and like aoid Bas Action and maybe the structure looks more like looks nicer you know the the internals are still really ugly and hard

to audit but that's the internals and you know maybe in 2024 will have better self-contained audited gadgets but that's like another question um but when you still do both you just went one step and you probably need to take a few more to actually gain some other properties that ziker rollups have around amortization which you you're not really going to get with that approach so cobe's point about verifiable FH and verifiable NPC I think on the positive side um uh I do feel like rollups end up being the place that people will experiment with those and somehow pretend that they don't need a token so

you know like I think that that'll be an interesting experiment and I I think it's sort of in the vein of like the singer saying like what if we move past optimistic or ZK well I think one of those things is like your best candidate all right so the next kind of overarching topic that I wanted to cover were research predictions specifically Nico you hinted at a few of these but I'm curious on the ZK front you know if we if we want to say let's just for Simplicity say 2022 lookups 2023 folding 2024 small Fields like what is it maybe small field I think multivariant polinomial iops so you know we've been

in univariate land for so long I think multivariant is we've seen a lot of it at the end of 2023 okay and I think going to see more in 2024 and again this idea of we no longer need to think about posting proofs on chain so we are happy with longer proofs but faster provs um that also comes from uh this sort of confidence we have that oh yeah don't worry worst case we'll put that in a proof right oh no no not just Hardware sometimes people are like oh we make these big proofs but don't worry I'll wrap it in a gr 16 proof I'm not sure it's that easy and I think maybe we're going to need people to look at that a bit closer and actually build these wrapping proofs and make them fast um so maybe that's something for

2024 Kobe I think in terms of research I would avoid for myself to make predictions on the lowlevel stuff but what I think we might see more from teams publishing is using these new methods of either using lookups in a very um extensive way or using um binary Fields uh or using small Fields or so on I think we will see more papers that exploit the composition of these things and more constructions that are using the composition of these things and potentially also addressing what niik

was saying around yes let's wrap things and let's make it happen um I I believe we'll at least see some teams making concr create progress around around those because that will be a big improve and performance so that's something that I would I would expect to see on the applied research site um and more concretely I think one of the things that we've seen in recent papers was that they've attached benchmarks they've shown some stuff but I'm not sure that these benchmarks are really applicable like they talk about table sizes they talk about about measures that don't

necessarily translate to what I'm going to to need to understand whether this fits my applications and I hope that we'll see more works that address that part as well and I think we've seen one of them which is called ZK bench which is nice but I I'm pretty sure that we'll see more of that I was wondering do you think that in 2024 a research paper about ZK from the salana team could come out is there any chance that salana gives a [ __ ] about CK in 2024 no no so so there are teams that will Implement ZK stuff no doubt do I expect new research from

them no I but I do think there'll be implementations I think it's like light protocol that actually is like sort of live is I think like you will see tons of stuff there um but not definitely not like research just people porting over code bases of other things I mean I think you'll see other other things appearing in salana land but there will be more there'll be less ZK I mean what does the dog with a hat need privacy for like that's the first question to ask I mean who knows maybe the crazy nft Market over there forces a need for ZK maybe the the the real ZK Dow voting I

know there's been experiments but maybe it's maybe it's there that it really takes off but that still wouldn't be research it wouldn't necessarily be like yeah news Z keep I I do think their their their Nosa safe alternative and dowo stuff is actually really good they actually formally verified it uh like it's actually the two I think largest formally fully formally verified uh code bases in crypto are are probably safe and squads and then of course a bunch of D5 protocols have formally verified a lot of the pieces but maybe the whole thing but these multi sigs that are Dow multi sigs have been like the most stress tested formally and I I I think that the salano one's actually probably

second to say it's like you know in terms of rigor so I I think on the engineering side you're going to see tons of stuff there's no going to be zero research all right I think we can move on to our next category which is dream applications o like at least the way I formulate this is it's it's sort of user facing applications something that like a an end user could potentially touch so does anyone want to go first on this one I I don't know what application yet but my dream for 2024 is to see like something outside blockchain of web 3 and something kind of tiny and inconspicuous you know something where people don't actually really know that they're using ZK but it kind of works and eventually they'll realize it's

necessary like I want that little seed to be planted cool don't know how um but I want that to happen so I had a few things on my list so one of these would be again like going back to to Identity work I would want to see um more ways for people to use their the rep to identities and basically use cryptographic signatures they already have um to to do things on chain that would be great um again I would want to see this in a seamless way like Nico said that would be really really awesome and as a consequence I hope that client side proving will improve because that will be something that has not been

thoroughly looked at um and that also connects to you know this topic of attested sensors and all of these Transformations that you do on on sign data and I would really want to see that explored more because that that might unlock ways to use it outside of of crypto which would be great um although there I also see intermediate ways of doing it with some cryptography and like some tees maybe but like that could be interesting as well um yeah and and in terms of what what else I would want to see and again this is just my my dream again

which is I want to see more uh experiments with with MPC so for example gaming infrastructure using MPC that would be really really amazing um because that would be this kind of non-financial privacy that that users can use again without feeling they're using something heavy or or weird and so yeah those are things that would be cool yeah I mean I have a a weirder higher level dream that's like maybe not an exact application but it's just like right now kind of what Kobe was gesturing at which is like ZK feels like this very heavy thing right you want to do something to go a server fwhere the middle ofwhere

consum a of electricity to get to you with you know a ni 256bit uh proof of whatever the hell you want um and what what I want to see actually and what would be really cool is um and this also gets a little bito thing which is you can do things on your computer you people just you press enter something gets shipped off and it's straight from your computer that's fully actually privacy preserving not this fake thing that we kind of do right now y uh that sits in someone's like 4090 GPU or whatever um and and it and your computer actually doesn't warm up too much when you do it right so it it functions as a as a cat friendly but at least Still Human friendly uh uh lap warmer so that that's that's the dream I

mean that's the high level uh silly dream uh of just like making things truly practical mhm I think the lower level but kind of interesting dream is Apple adds a Ted photos to an iPhone o and I think it'll I think it'll happen I mean there's no reason why right now Apple can't just sign like you know a picture you take from your iPhone or something right they already have a bunch of things like all their parts are DRM and all this crap which I I'm I'm not going to comment whether that's good or not um but you they essentially right now the the company most well uh posed to do this kinds of things is actually

Apple um because they have such tight control over what things exist in their phones and things like that uh and I I'm pretty sure using their like fancy uh whatever the t2 chip or whatever actually I it's a different one but they should be able to do this kind of thing quite quite easily actually isn't this bad for worldcoin if that happens because then anyone and their mom will just make worldcoin as an app on their ZK chain and you don't need the orb good um I have no idea I mean I I don't think you can do General you can do a lot of stuff but you can't do General computation on their current chip I just need the attestation and then I can basically replicate most of the world ID we need the a St yeah the question is like what do they do at a station 4 is

it just images or is it like scans or something you know I was thinking just images because I I think other things will probably require a pretty nice user accessible API um but for images I mean if you just sign a thing with like one of the keys that Apple has you know and you go check apple.com you know Pub keys or something you can be like cool yeah like this image actually was signed and this key hasn't been revoked so actually we're we're cool this was taken on an iPhone that would be very interesting yes but by the way I think Apple is is positioned to do it not only because because of its tight control but also because it has shown that it is that they're interested in making changes in that direction for

example end to end encryption in iCloud is kind of mind-blowing to me the fact of yes that it works well and it it is actually good like the design is great so same with time machine if you've ever used time machine it's insane like but in what you you're describing there that's the attestation it's not a zkp in it in like oh but then you can start proving that the editing you did yeah on this aasted photo but you're not saying Apple would provide that like cuz like wouldn't it be really cool if Apple had the ATT testation and the filters that also like and maybe they would have that where they're using a zkp through any sort of filtering any sort of like post

production and actually spit out something that's like proven back to the source yeah they might acquire someone who does it I think the simplest thing right now that's going to happen is like you just provide the original that's fine and then you just provide the edits and someone can be like oh yeah like those are actually not that different or something yeah you know yeah um I'm sure at some point actually they could even do it all in Te or something I don't know if they're going to do this but uh or they could do The Vision Pro has a good te or like Enclave like it has yeah it has like a much better Enclave than the phone so I'm actually more curious about the Vision Pro as like the the thing that is easier to do this with than the and I think it comes out in

like two weeks or something now so what is the Vision Pro is it a video camera the VR the mix mix mix reality goggles yeah apple it has a much it has much better like set of processing sorry I think you meant the spatial Computing device sorry sorry I I I forgot marketing speak 101 my bad you're the CMO tun no no I'm the CMO for this podcast I'm not I'm not the CMO for Apple CMO for apple is like above my pay [Laughter] grade actually back to your firstborn gear like I had something similar written down which is ZK creds which is like this this paper from two three years ago and it's basically you and

your government ID at home making that into an anonymous credential you're at home you're using standard Hardware there's no trusted party you interact with a blockchain maybe a bit but that's it you transform your government issued ID into an anonymous credential that you can use anywhere you want and that's on my like dream list for 2024 like and I think teams are building towards that and that feeds into the sort of ZK email ZK login type of thing it's also the same kind of reasoning going on yeah and and I think pulling on that thread uh things being kind of as decay and being extensible will will be an increasing theme which would be great I would love to see that tun do you have a

dream application I mean mine are always the boring ones like you know using it in an me auction instead of a te using you know like basically take all the [ __ ] people do at tees and find one of those [ __ ] things and like do it with ZK or IO or whatever you know like in the case of like signal right signal is a great example someone who uses [ __ ] Oram and sgx and could do things with ZK but literally has said it would cost them a 100 times as much and their ux sucks and they tried and they gave up and I feel like I feel like those Ed cases are very tangible and like here today right like signal signal they actually wrote a paper with Kobe about having an alternative for their like sgx

o around thing really I just think those things are so those things are so easy like those things are exist already and they're like are they're more like engineering plus optimization than they are like you need a new prover but no one finds them sexy right everyone would rather like they get their 10% better folding mechanism out instead of working on that I I mean I have more criticisms for Signal more generally in terms of how they use cryptography right like even the their use of cryptography is not just they're not using ZK it's it's more that they have this whole double ratchet thing and that can give you OTR of the record so that you can effectively deny any message that you've

sent but you don't have any tool to do that so that that's why people can still see a screenshot from signal and believe it and that shouldn't be the case like if I see a screenshot from signal it should be completely fake like maybe when I send the message it should just be garbage right after don't tell the US government that in their prosecution of s SPF where 50% the things were signal screenshots deniability wasn't higher up in your dreams list um I I think it should exist but I think it I I think that I would really want it to happen and actually you make a good point I

want it to happen because I kind of feel that it replicates the in-person experience when you're online because when you're talking in person you have this confidence that that you if if you will be repeated no one will believe them or something like that and that should Cas list to us don't worry this isn't your confidence could be yeah right now I I know but like teron said they are using this as evidence in courts and they should yeah exactly exactly should I I I I just think like there's there's a sense in which there's all this stuff that already exists and don't get me wrong despite all the hacks to he and whatever they clearly they are being

used by hundreds of millions of people every day maybe maybe billions right and I don't know chipping off 1% of that to me is like the ultimate that would be a good success case right for ZK I mean every every Apple machine there you go has a nice little te in it actually multiple if you're not careful right yeah my dream applications were an identity product people want so I know you guys already hit that uh a popular privacy app that off a way to have privacy but does manage to keep some of the criminal activity like clearly at Bay provably at Bay and sort of this is a wish that I mean this is this is a use case or application that's been floated

so much and I doubt it's close but it would be like any genetic testing like I just the more I learn about the fact that that's out there and there's so such poor privacy controls it's insane it's so dangerous and so any sort of medical like sensitive Medical Data being secured by ZK would be amazing this is Anna discovers uh differential privacy and for databases uh Arc this is like a thing that a bunch of people were doing back in 2010 10 years ago C 2016 or something which is like how do you do like genetic testing privately without like without too many queries and all

this stuff I think a very funny part about this is 23 and me wrote a bunch of paper about this and then proceeded to never use awful not give one [ __ ] like not at all I think the Stark paper like that was the example that they gave right like DNA they know that actually my last stream application is also the uh ZK outside of blockchain there's no T tangible use case I can offer there but just something the signal the signal use case is a is a non-blockchain use case okay my one thing is ZK and blockchain just happen to play very nicely because blockchain is a very nice database to record public things that can then later be verified it's also a very good public

verifier so I write proof that is publicly verifiable who do I get to verify it blockchain and it has it has a funding mechanism built in I think that's a big reason why definitely true but it's just like you know if you're going to go publish you're going to go down this Rabbit Hole of doing a bunch of privacy stuff right and then be like okay cool I'm just going to host like my public in like this random server that anyone can take over it's like okay well like why the hell did you do all of this damn thing in the first place someone could just trust your server you know to do the right thing so it's kind of this very nice like okay I am publicly broadcas saying that for example this Pub Key is indeed mine and like here is like a bunch of fancy stuff like and and

now anyone can ACC it at any point in time like without being dos because that's kind of the whole point right so uh anyway sorry that's my soap box over although I think more generally and um that blockchains like you said are a good public database that you know you can put things there but more generally blockchain has been a good source of signed data and now that you will have more signed data you can do more outside of blockchain all right we have 15 minutes left in this episode and we have a few kind of I'm thinking almost like rapid fire questions we can go through because yeah I had this list that I wanted to share it was best case

scenario for ZK worst casee scenario for ZK most likely scenario for ZK Black Swan event which I love that we try to predict those and breakthroughs that could change everything okay so best case scenario for ZK in 2024 go something that isn't a rollup gets more than 10 users very cool and isn't the orb okay yes I want to see some kind of solution into like the Civil attacks right so again not the orb not necessarily proof of personhood but get use DK to get rid of like spam and Bots essentially Kobe um non-financial privacy gets more

attention and the performance of client side proofs like GMO saying that that's going to be really important for 2024 okay best case GMO um uh a famous web 2 application using ZK and some some small but nonrival way and this is stealing a little bit ofos wait quick question quick quick question do you consider non-consumer facing web 2 applications as yes I do actually so like if Cloud flare does this and no one knows that counts this is best case scenario for ZK I mean mine was breakthrough into the mainstream conscience in the best way in like a really positive obvious like the benefits are clearly articulated way that would be the best case scenario for

ZK in 2024 we're going for some real dreams here yeah yeah okay worst case scenario for the ZK I'm going to start because I just shared mine mainstreamed in the worst possible way connected to Crime yes or whatever like something worse like the worst I feel like that was 2022 it was a little bit it's happen already yeah but I feel like it could be worse I don't know if it was that mainstreamed like no no I agree it wasn't a lot of people outside of this space didn't really hear that much about that it was main crypto hearing about Z they didn't associate it necessarily with ZK but that's my worst case scenario for ZK wait I have I have I

have another one to add for best case oh okay okay sorry best case there's a meme coin based off ZK we need we need the Doge coin of ZK okay worst case there's a meme coin of ZK just kidding just kidding but that's good okay okay worst case though tun what's your worst case the worst case to me is probably ZK rollups uh are unable to get traction at the expense of optimistic Roll-Ups and then optimistic rollups sort of to Kobe's what Kobe hinted at in the beginning like maybe don't care as much about full guarantees over time uh and don't you know like the the the saying like Hey we're going to

add ZK is sort of more lip service and action or something like that would be bad right because like it would kind of be like Str ZK gets strung along as like ZK rollups or ZK used in rollup security is touted as being important but then it's sort of more of the straw man thing and that's not like good in the long run because then people will just be like well no one cares about this technology right so that to me is the worst case doomsday that's a good worst case Okay GMO what's your worst Cas worst case is cryptography Wars 2.0 that's my worst Cas what does that what does that mean uh so I guess quickly in the '90s there was a bunch of crypto cryptography Wars because uh the

US government did not allow uh the creation or export of essentially open source software to do cryptography it was considered military or something right it was that's right and so putting it in open source would be considered like you know releasing militarys to like foreign governments or something the nice thing is that so many people working in ZK aren't in the US so I don't think this is like this is much less of the '90s and ' 80s problem as it is than it was unfortunately then it was like 100% here yeah but but the EU and the UK are not headed let's say in the in the best Direction on that exactly yeah as long as there's enough difference in Direction like as long as they're not like a vector all pointing in the same direction but there's some diffusion I actually think that's fine

then they won't agree on something you can just jump them from one to the other as needed exactly exactly okay Nico uh worst case is we get cold feet like we haven't found the killer app we started being a bit scared VC money is dwindling a bit cuz they got excited about llms and people stop dreaming and building right like that's for me the worst case that's a good worst case yeah it's great okay Kobe uh so first of all the the constant one is a big hack um snarks are broken Starks are broken beus is broken that's my most likely that's my Black Swan most likely oh my God Nico well not in that way I I'll tell you in a second when we get there that was my Black Swan

also wow we all have the same Black Swan but uh two orders of magnitude in in performance to fix it and we we kind of go back and be really sad and maybe op to use centralized or optimistic solutions for many many years because of that um that that would be that would be bad that's bad but yeah okay most likely so I have hack but either way like snarks are broken but more like one ZK app or one thing that is deployed is going to have vulnerability and we're not going to catch it before it's life and we've seen like audits catch really important

things we've seen white hats and Bug Bounty programs catch very important things so I think there will be a hack and the sooner the better if anything wow you're the person who exposes their children to allergens to reduce their future allergies yes a good idea mind you yes perhaps unclear but okay most likely honestly for me it's most likely the St remains like a mostly research year with not that many applications found o I know sorry tun most likely um most likely is that I'm gonna I'm going to be The Optimist here is that one ZK rollup I'm not sure which one will cross a billion

dollars in some notion of value locked this year but the irony is it might not be a ethereum ZK roll that's that's like I'm not saying that that I I think that's true but I do think there's so many non evm ZK rollups that people are interested in in 2024 and like I think one of them will launch and maybe we'll have a craze around it also is it zik rollup if it's not on ethereum don't don't don't talk to any of these Bitcoin ZK rollup people you'll suddenly you'll suddenly get some very funny answers oh is that where you're you're thinking it might be on bit no no no no I'm not I'm not saying it will be

but I I I do I do think that there are the the Bitcoin ZK rollup Community is very weird and I do think if they launched the billion dollars of tvls are easy for them to get compared we should have that on the show we have never covered that that is like is it weird good it's well I do feel like they're forced to use Starks but that's something you know maybe on an episode where you interview someone you can talk to them about but like I think it's basically impossible to do snarks and mainly cuz litic curves are very annoying to deal with in Bitcoin script okay versus hashing Kobe most likely um I think some specific zik applications will be deployed on on new systems or like in a

more performant way but when I say applications I don't mean something that really people need or use yet but I just mean use cases that are possible for people to use so something like zkl or thing of that sort I think that kind of thing will be deployed and I also think that um we will still have messy ZK developer experience that will not be what we want yet although there is huge leaps there but not yet and the last thing would be that we will potentially because of these new things that I mentioned there will be

more devs doing ZK don't we love their messy experience like it's part of the charm of doing ZK isly that's part of the fun right all right Black Swan time it sounds like we all have the same one well yeah one of mine was Major bug in an underlying gra 16 library or some fundamental plunk version that everyone is using yeah uh same and it's funny because like but it's funny that if we're all convinced of this is it really a Black Swan event or is it just that yes just all waiting to hit the wall yeah black Swan's kind of a funny it's like the thing we fear maybe there should be like Boogeyman like the the thing under the bed the thing we're actually scared of but but it exists

that's the thing yeah I I maybe maybe I would condition mine on like I actually think the source of this will like more likely come from translation rather than like intrinsic bug like you know the there's all these compilation steps and translation steps and like actually auditing the whole end to end thing is much harder than any individual piece and I think those translation steps like between like a circuit and a polinomial representation or between you know somewhere in there yeah that's exactly it's like stuff like that that is much more likely to be the wow mhm there's a really good report by the open Zeppelin

team actually about a recent like sort of Mis implementation of fimir bug that they found which I recommend G Mo Black Swan this is the one that I had the hardest time with I mean like yeah I could say hack but um I want to come with something more interesting unfortunately I don't think I did I the hard part of black SP that you can't predict them I have another one should I share mine then yeah go for it I'll fill in so it's a politician a powerful politician hears about zek K doesn't understand it and trains their guns on it I guess that's like guo's that's sort of like gmo's worst case it's yeah it's we're was going with it with that like it' be the unfortunate happen stance of

someone who doesn't really know what they're talking about doing things but that that many such cases I guess yeah I think another Black Swan would be that a politician is a proponent that's even more surprising yeah black have to be bad right I guess it's true yeah it could be it could be a like bad usually imagine if like Donald Trump was like I love ZK no one can see my nft sales what are you gonna say what are you gonna say about that like oh my God very good okay last point we don't have too much time left uh I had no answer for this one it's breakthroughs that could change

everything I have some some cryptographic dreams yeah that's what I thought there might be some like cryptography breakthroughs that could change everything so one of them is what we call witness encryption so I encrypt something to like a public key and the private key is like some secret that no one knows yet but someone might Discover it later and whoever discovers the secret first gets to decrypt whatever I wrote that has some really strong potential applications the other one is uh indistinguishability opusc like dis disability indistinguishability obfuscation that was what I was going to say damn it oh really great happy we're

all here uh but it's the idea that like you can hide the inner workings of your program so you can have a program that's public but you hide the inner workings so inside it can contain like a private key and it can do some secret stuff um but the rest of the world isn't aware of how it's happening cool tun do you want to expand yeah I I think uh IO at least has like some of the dreams that people have of private computation um despite sort of all the complexity Theory problems with it of like Well if you really can make a practical IO then you have some very weird things about p p equals NP or not have have to be true

um but I I think there's kind of this idea that you know if we think about something like ZK you have like fully public program fully public Trace can be verified and then you have FH where it's like there's still some public information uh sufficiently enough public information that side Channel stuff you have to be much more careful with but IO somehow lets you not have to think that hard about that in theory so it's this idea of like you can be lazy about how you send people information and still be kind of like all right well they can't really figure out everything versus like you know whenever people talk about FHA right the number one criticism which usually te people

love giving is uh oh well someone has to decrypt the the final output right and like the threshold group has to decrypt it and and IO lets you avoid some of that and so the te people have no remark to iio but then again IO also means complexity Theory stuff is not fun if true so that's interesting that's actually not my biggest criticism with FH biggest you should read Andrew Miller's criticisms they're always about this yeah like my my biggest is the fact that like you don't know what runs there like people can run anything about in your server text so what is a breakthrough that you think could change everything so unfortunately it's similar to what the others have said a bit which

is this indistinguishability ofation I we've just learned short form okay IO and lowercase i capital o oh um where the way that I like to look at it and I think that I've seen also vitalic using this framing which is that smart contracts can then hold private keys and which is something that would be extremely powerful that that would change a lot amazing what about you GMO what is a breakthrough that could change everything uh I mean I I was going to say like actually practical like Z like I think I think that's still the biggest one it's like it's boring it's simple it's easy it's something we could

probably just do but still haven't done all right so I feel like we have covered all of the questions that I came prepared with all of our predictions but we actually also have some like tangible events coming up that we wanted to share so next week January 16th we are kicking off our 4-we Z hack 4 online event and actually it will be kicking off with Kobe Nico and I doing a presentation this is sort of Nico's brainchild the ZK jargon decoder but we're doing a session live uh as part of our event so maybe we can talk a little bit about that yeah

yeah well I started when I joined like this ZK world I guess like year and a half two years ago um came across a lot a lot of new words and often you get different definitions for them and eventually when you understand the concept you're like ah that wasn't so hard why did it take me so long so I started writing these small definitions which I hoped were shorten the path for everyone and uh I guess now we're doing a live session so we're going to cover some definitions we're going to hopefully take some live questions as they come in yeah should be fun and this is happening on January 16th we are doing four weeks this time for zck 4 online it's virtual anyone can join it's free I'll add the link in the show notes

we might even have a mention of this in the intro of this episode and I want to do a little throwback to the origin story here because Z hack generally as a project started as this multi-week online event and this is the fourth time that we're that we running it anyone who's familiar with the show would have heard about ziki hack it has since spun out into a full like its own project its own company um it has a very active Discord and um yeah this time around we're going to be doing three puzzles Kobe do you want to share a little bit about the puzzles cuz I know they're coming together yes already so yeah share a little about what people can expect yeah happily so just as a context for people that maybe haven't seen all

the puzzles from before so usually the puzzles are about how protocols can break in ways that are real and some of the puzzles that we show are actually exploits that have been out there in the wild and what I know about myself and I think I've seen that works for others as well is that people really learn the best by doing and so when they understand how to actually exploit the protocol themselves using this more self-contained example and that helps them learn a lot on the way um and specifically for ZK hack 4 and we came up with a theme for the

puzzles which is what happens when you handwave your security proofs um which is which is really really something that we sometimes see out there don't roll your own crypto and if you do write proper proofs at least um and when you don't that's what happens and uh yeah the puzzles are as usual are are Works based and people can catch up I'll also share some resources on X for people to to catch up on things um but uh yeah that's the plan very cool and our plan and and since this event is

actually running four weeks so we're meeting every Tuesday for a workshop and in between those workshops that's when that puzzle hacking competition is running so we're going to have three unique puzzles spread out over four weeks each week we're g to let one out in the wild and then hopefully you solve it quickly it's very CTF like So the faster you solve it the more points you get but obviously it has to be solved correctly um and there will be prizes and glory to be achieved through this we also I mean Kobe we this is our fourth time doing this the the ZK hack project really started from your idea of doing the puzzles and then we kind of built everything around it I know for me it's

like a way to get to know some of the expert in the space especially those who might be like working in teams but aren't on stages they're not allow on Twitter all of a sudden like the ZK hack puzzle hacking competition brings them out and then we get to find out who they are ZK hack one has a special place in my heart cuz that's that is when I first got in touch with this whole Community wow really like I had studied cryptography and ZK at University oh no way and then I don't know didn't really find anything that was kind of privacy oriented with good people until I came across ZK hack one wa I didn't know that Nico that's so cool yeah nice yeah yeah I I met a lot of really smart and

quieter people like you said through ZK hack and my my impression is that it also helped these people get get into like the ZK job market as well which was great to see I really LED seeing that and I also love to see how it grew to have these study clubs or study groups that uh that exist now cuz I think that's also been really influential for people totally you just mentioned actually hiring we should mention there's going to be a ZK jobs Fair which is a recruiting and like job finding opportunity we do this virtual we kind of create a virtual space teams can meet with potential applicants but

it's very casual so you can do this without like before you send the application get to know the team you're thinking of working with um at the job fair and that will be happening after our final session on February 6th cool okay I have one other event I want to mention which is zk1 applications to attend or speak are open now I'll add the link to that in the show Notes too it's invite only so you do have to apply even to attend so yeah keep that in mind tun you have an event coming up too what's going on in your world yeah so just High LEL um basically a bunch of of us sometime last year I think around e Denver last year were thinking about

sort of the state of more uh like defi research and mechanism research within crypto um there's sort of this interesting thing where it's sort of in this like no man's land in between econ pure econ research and pure sort of algorithmic Game Theory research um and it doesn't hasn't had like its own conference venue or place where people kind of share ideas on it you know kind of it's not really in consensus it's not really in CK it's not really in mechanism design it's not really in econ like it kind of has this like weird it's in the middle thing and so we were like Hey how do we make a new venue for research that is in that form and like some of that has gotten published in all these different subfields but like a

place where that that the real gation of those ideas can can come to their full fruition and that is what tldr is which stands for the latest in defi research it's not maybe the name is a little bit not sounds amazing and we're going our first conference in May we also funded forget the exact number but I think it's like 12 fellows who for doing re research so we we have a fellowship program um so people who are interested in defi research but like needed sort of the a kick and some money um kind of can can do it so that'll be in May 2024 amazing thank you all for coming on and sharing with us this uh look forward at 2024 in CK remember everything you have

said in this podcast Canon will be used against you in next year's podcast true all right uh big thank you to the podcast team Henrik Rachel Tanya and to our listeners thanks for [Music] listening