Episode 279 - Intro to zkpod.ai with Anna and Kobi

Transcript

foreign [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 [Music] so in our last episode together Kobe and I had shared a recap of the state of CK and had planned on speaking about zkpod.ai but we barely got a chance to do this since we had both run out of time on the call so this week we continue where we were introducing you

to the zkpod.ai project and sharing what the motivations and Inspirations were for it Kobe shares insights about the tooling available how he used them as well as their limitations and then we share a few Clips we chat about our findings and we talk about the larger Concepts that this brings up the accelerationist versus doomerism topic does make an appearance so hope you enjoy and also get in touch if you want to help out with this project we are still working out what it will be like but we'd love to hear from you if you have some ideas or tools that you're aware of that might actually improve some of the things we're doing if you haven't yet please do check out ckpod.ai to get a sense for what we're talking about now Tanya will share a little bit about this week's sponsor

[Music] anoma's first fractal instance namata is launching soon namata is a proof-of-stake L1 for interchain acid agnostic privacy namata natively interoperates with fast finality chains via IBC and with ethereum via a trustless two-way bridge for privacy namata deploys an upgraded version of the multi-acid shielded pool circuit otherwise known as masp which allows all assets fungible and non-fungible to share a common shielded set this removes the size limits of the anonymity set and provides the best privacy guarantees possible for every user in the multi-chain the masp circuit's latest update enables shielded set rewards directly in the shielded set a novel feature that funds privacy as a public

good follow namata on Twitter at namata to learn more and join their community on Discord discord.gg forward slash nomada so thanks again anoma zero knowledge is changing the world and until now building ZK applications meant learning new chain-specific languages and complex cryptography but not anymore with snarky JS the easiest use of ZK SDK developers can add the magic of ZK to their apps using typescript whether you're targeting Mina the leading ZK blockchain or off-chain applications snarky Js from o of one Labs has you covered with support from infinite recursion in browse improving and so much more the full power of ZK is

available to everyone visit snarkyjs.01 labs.org to get started you can also find the link in our show notes and now here's our episode [Music] so Kobe and I are back once again to continue the story we started last time but didn't get a quite a chance to finish hey Kobe hello we're going to be doing the episode that we've been meaning to do for a while all about zkpod.ai so let's start off with an introduction to what is zkpod.ai and how does this have anything to do with this podcast [Music] data ask questions about it and then

have the answers talked back to you in Anna's or mind's voice cool yeah so from a user perspective what does it look like to them they enter a website they write a question and then they wait about minutes and then they get an answer cool and actually We'll add the link to this obviously in the show notes we definitely recommend you check it out if you haven't already you can ask it there's a prompt question there but you could also put in your own question and you'll see how it how it works it sometimes it has a good answer sometimes that's good but I think you should play with it what we want to talk about today is sort of this introduction to the project what was the motivation inspiration and some of the tools that

you've already used as well as some of their limitations and then we can talk about some general future Concepts future ideas but first let's talk about this inspiration so what got you interested in doing this what was sort of the starting point for the zkpod.ai project the inspiration came from me first of all wanted to try all these new technologies around GPT and llms and all that but they found someone that has been doing a lot of experiments is called Yohe and he showed a bunch of interesting experiments that he's been doing with llms to create different kinds of pipelines like questions and answers or agents that can perform tests

and things like that and the basic tool chain that he used was something that's called langjang and langchin has been growing extensively over the last few months it's a very all-inclusive tool chain that allows you to process data to query gpt3 and to basically create all of these very complex and long pipelines that extend beyond the simple apis that openai gives you what is Lang chain like what what is it is it just like a library which has that people are adding things to does it tap into the chat GPT libraries and like the the data that it

brings or is it allowing you to put your own data into it Lang Chain by itself or more correctly open AI by itself it gives you a few a few different apis so it gives you an API that allows you to ask questions so or more correctly to do completions you give it a partial text and you know how these models work the these Transformer models these LMS they are trying to predict water is the correct next word that should appear when you have a text in front of you so you give it a text and then you say please complete this text and you know give me at most 200 words

and then you will finally have some completion so that was the API that was common to use back then that's gpt3 mm-hmm and later on you had you had more focus on chat models so GPT three and a half GPT and gpt4 so jgpt is using underneath the chat model like gpt3 or jbt4 and in that model you're actually asking it to perform like a chat so you have a user you have an assistant which is the AI you also Define the context of the system which says what is the behavior generally of the assistant and then you just give us the transcript of the chat

and it completes a new message so it can take a sequence of messages which is a discussion and then give you the next message that they they I would say when it sees this message but that's open AI right yeah that's the only thing that it gives you okay and this is stateless not only that it's stateless it is limited what do you mean by stateless though what what does stateless mean so it doesn't learn okay well whenever you're fitted with data and you call it next time with some other question it's not like the GPT interface where you have the UI where you can have some persistent session that goes on and on and on and on yeah yeah it basically

forgets each time when you ask it the question right and that's the statelessness feature of it or statelessness aspect of it and it's also Limited because it has a maximum number of words or as they call it tokens that you can put in the text or in the API when you call it so you can put a long discussion you can put a long text so usually you can't put the whole paper or article I've tried it with some good articles from people that have been on the podcast and they like to write a lot so it doesn't fit in okay and that's where the library is coming so libraries like land chain

they allow you both to interface with apis like open Ai and others but let's focus on open AI but they also allow you to integrate complex pipelines so it allows you to integrate the pipeline that splits documents and then sends its chunk of the document into the AI and create a summary let's say a sentence out of each segment which could be a few paragraphs and eventually you have a bunch of sentences and then it can also summarize that so you can build these pipelines that create a summary of a very long document using this library and then it also has

State I guess does it have a memory or you can you can you program it to have memory yeah so that that's also another module that they have they have let's say conversational memory so if you have a chat the chat is also limited because you you have to cap the discussion so what it knows to do it has one of these modules I think it's called conversational memory after it reaches the limit of the amount of words you can input it summarizes the entire conversations that went on up until now and uses it as a context and then puts the new questions after that context I see so it doesn't have to hold in its head the entire conversation

it can shorten it summarize it and move forward but when you talk about sort of the limits where are those limits like text limits so they come in a few places they come in the the how the model was trained I believe but they also come at the performance so for example gpt4 that supports very high number of tokens and say 8 000 tokens and there are even versions that support 32 000. then they start to be slower they cost a lot and it's it's more complex to create these models generally what's a token is that text length like what's the equation there so I don't really I don't know

exactly how they translate but the Benchmark that you have it's like parts of words it's more like the significant part of the word or more like syllables but the Benchmark that you have is usually that it's three three quarters so if you have 8 000 tokens it's three quarters of eight thousand that would be the amount of course you can put oh I see okay so you have less words than tokens or more words than tokens okay so Lang chain comes along and you see that I want to get back to our story um and you're like wait a second I can do stuff with this yeah I know a place that could be used

with questions and answers that you know some someone that's called Anna she's asking questions and people give answers and sometimes someone like Kobe who's also on the show is asking questions and getting answers yeah that's cool and I really wanted to experiment with that so that's how I started looking into this I I got luncheon I integrated very productively I already had transcripts available from the ZK podcast website right like that's that's one of the best features that the zika polka says and I know some people even just read the transcripts I think Guillermo said that right yeah little did they did they know how how much work we need to put into

transcripts although it's getting there but yeah yeah yeah yeah I know exactly and these transcripts are really good because they're not just processed by some system you you put a lot of effort into it yeah and they're they're fixed and they're polished and they're really good so well okay I feel like for this experiment we've learned that actually our transcripts could use some help turns out there's lots of typos and some of the more complicated technical words have been missed but we are working on that yeah actually we can talk about that later and as you can guess the transcripts are too long to to put inside a single prompt or a single call to open AI so what I have to do is to start cutting this transcripts into

chunks so that's something that LinkedIn allowed me to do and now that I have all of the transcripts of the entire ZK podcast it's like 200 of them well not 200 because some episodes in the beginning didn't have them but roughly now I can start asking questions and but that that felt like something that I've seen some people do already so I wanted to put a twist into it and I said you know AI is not only good for language like large language models that's not the only thing that it can do it can also produce voice so I look for the best voice cloning technology out there and I found it and that's when I

started sending you recordings of yourself saying things you never said freaking me out yes that sort of and that's where the ZK pod sort of experiment comes together into something that is shareable um when I first saw it I was had this mixture of Amazement excitement and fear I wrote about it in a blog post which I will add to this episode if anyone wants to read that and some of the feelings it invoked I think you and I coordinated this a little bit so like you had shared it with me at the beginning of March and I think we released it into the world about two weeks later yeah and during that time I I mean I did go through some you know back and forth just being like wow this

is you know one I was like well does this replace me do I is this the end of you know the need for a podcast voice at all we'd had a kind of conversation about where it could go but I think we wanted to first release it into the wild as it was and maybe we can talk a little bit about that so like we let it out you shared a tweet I shared my blog post we've since seen what people did with it like I think at first it was obviously just like we had some prompt questions that people just would just ask over and over again like the amount of times we've seen the question what is hyperplunk based on asked is it crazy because that's one of the the sample questions but we also started to see some very interesting questions people

starting to experiment with this and also try to push it to figure out like how much does this know what can it dig up and it's been really cool to see that yeah like it was extremely cool to see how it processes data from the let's say different episodes or different sources where it can summarize and create bullet points about the topic which is pretty cool now I think it's worth it for the audience to hear this if you haven't already played with the kpod.ai and asked a question and we're going to share two clips one is a pretty high level question and the other is very technical and so here it is the challenge of working with snarks is that they involve Knowledge from various

technical areas like computational complexity cryptography and applied cryptography this makes it difficult to train students and bring in Talent who are interested in understanding this area as it requires a comprehensive initiation process now here's the more technical example arithmetization is the step of taking a program written in some language and describing it using a set of polynomial equations so that proving systems like snarks or Starks can understand improve the execution of the program different proving systems use different arithmetizations such as r1cs for snarks and algebraic intermediate representation air for Starks you can hear that it sounds just like you at least relatively right to someone

who's listening closely there is a difference slightly in my accent yeah I think like the voice calling technology that I use is tuned to North American accents so it tries to reproduce that and yours is closer to that yeah I think so and if you want to try Kobe's voice we should do a quick sample of that as well so here's Kobe answering a question ZK cannot be used to build a house directly as it is a cryptographic technique rather than a construction method however ZK techniques can be used in the management of funds or data privacy related to real estate transactions and here you can see for some reason Kobe has been made British

no no idea how that happened I don't sound British at all I can barely pronounce English that's sort of like the story to the point of release we started to see the experiments and I mean some of the things I noticed right off the bat was like certain pronunciations of technical terms this has not fully achieved and that's I think more in the in the language part right that's in the voice The Voice can't read or pronounce certain things correctly and other times what we did notice was some things were just like plain wrong and there at first I really I mean I started to be like it must be the transcripts we know we actually do have typos mistakes in our transcripts

especially the older ones like just to give some story like some background for folks who aren't aware we have transcripts on pretty much every episode but the way that we're generating those is through like um what is it an audio to text generator and then we have human editor going over it twice but we haven't always had like a super technical editor looking at it and so there there's times where you have I think all the technicals in line and there's others where it's a little bit off it's been highlighted to us very much through this project because all of a sudden we're seeing it output things which are like a little bit off or they're not you know the spelling of a technical thing is coming out even through these answers sort of odd so what's kcg you know

I started to I was like oh my God that's totally me I'm just totally saying it too fast Casey G Casey G yeah anyway but yeah I think transcripts and like the quality of our transcripts all of a sudden became very very visually I was like okay this is something we need to fix and we're working on that right now we're looking actually into ways that maybe even some of these tools could help us make these transcripts even better and I think that well what we discovered is that that was also the point when they discovered that you know Langston is it's very good for getting started you know it gives you all these very neat pipelines and ready-made prompts that says things like you are a bot a helpful bot that wants

to answer questions and this is a question and using these pieces of context you will produce an answer answer colon and then you'll produce the answer right but that's always the best because in our case we wanted concise answers we wanted it to not respond if he doesn't know like we wanted it to optimize to saying I don't know if he doesn't know and the prompt that we use has evolved very very much since we released it and that's one of the things that made me use launching less and less up until the point where we just call the apis directly now

um and because they wanted full stimulation of the prompt process but yeah like you if you want more control you sometimes have to rip these libraries out and we see that all across software engineering right it's not special here yeah but that that was where it stumbled upon it here so interesting when it comes to the pronunciation problem do you have any ideas on how we can even fix that though oh yeah I have ideas okay so the the software that we use right now to do the The Voice cloning and the text-to-speech generally it's tuned to task of text to speech very directly there are some ways that you can help it pronounce correctly

so there's all this phonetic ways to write words that you see in Wikipedia for example so those are ways that you can help it but there are also other methods or other voice models that have been either released lately or been made into production quality libraries lately that are even more expressive so it doesn't only know to pronounce a sentence or to say sentence out loud it also knows to do things like laughing and giggle or whatever it can do all sorts of stuff so I have a sense that those will be more flexible to make

better pronunciations interesting like would it ever make sense though for us to also like re-record some of these things because we're like why can it not I guess because it's two different processes right so you don't you don't like you're feeding it my language and then you're giving it the text to read but you're not feeding it saying those words right no it doesn't learn that that's why I think I've heard it say stuff like ZK snarks is sometimes like the snarks or something yes or there's like these weird like yeah nobody says it like that but yeah but yeah it doesn't learn it just learns the the shape of your voice and then it produces whatever you give it it doesn't

learn specific chords one other thing and obviously as we're talking through this project I really recommend folks play with it like try it out in these different ways get it to say sort of ask it questions about some of those like maybe harder to pronounce words and see what happens another thing that we found out we could do and like this was definitely true people testing it was you could just ask me a question or Kobe a question in a different language so you could ask in French German Mandarin you can actually ask in all these different languages and you'll have my voice attempting to answer those questions apparently the texts are inter like mostly right the pronunciation is hilarious because it's basically like an

English voice poorly speaking other languages which I learned when I heard the French one which I actually speak and was like Hey I can do that better than this but you know the software that we use now supports multi-language stuff okay we've never tried it it's like where they released it two weeks ago should I record myself speaking French and then see if it picks it up yeah let's do it okay okay we're gonna make the French better then maybe even by the time this comes out and then maybe it can produce yeah multiple languages it's gonna be fun cool maybe one important step in the process that I didn't mention before is how we actually find the right places in the episodes you know you ask a question and then you

want to find the correct chance because we can't ingest whole transcripts so we go over the 200 transports and then we want to efficiently find all the correct chunks that are related to this and specifically we split all the transcripts by speaker so we have you speaking and then the guest and then here and then the guest and all these individual paragraphs call introduce different chunks and there is a way to encode these chunks in a way that is efficient to search close chunks to them so there is this process called embedding where you transform a sentence or a group of words or a paragraph into vector and numerical

vector and then what you can do is you can find other vectors that are similar to it that are close to it in terms of distance and that's one of the main methods to use in order to find the relevant chunks that are related to a question and we pull like 12 of them we put all of them inside our prompt and then we ask it to produce an answer out of it so that's kind of how it works in the back end that's so interesting I actually didn't know that yeah you know we use the open AI model to do it they also have a model to do that like to do embedding and it produces a pretty long Vector it like has more than a thousand places and we also optimized that process a lot like

we started with something like a database that is file based and we didn't have the ability to add new episodes which kind of sucks so we had to process it from scratch each time so it took like 15 minutes 20 minutes each time there was a new episode which was kind of annoying but we moved to a database that is really cool that I use now that's called leviate and it supports all of these Vector operations internally so it even does all of that for you which is really cool and optimizing that process of retrieving the right chunks was a challenge and this allowed us to

improve the quality of the answers quite a bit so for example we don't only do this Vector Research Now we also do a keyword search which is like a traditional keyword search that we know from Sandwich engines so we combine all of these different chunks that we get from different places using the keyword search Vector search and together we produce an answer so this pipeline I think is under discussed people don't stumble upon that when they start out and when they get to the part where they do want to do it it's suddenly a big leap okay like in terms of the amount of software you need to install the amount of work that you need to do to

pre-process data it's not just calling an API with a completion anymore and so the library is like language and others help with that but that's when things start to get real and you say okay I have some work to do is this what you just described is that the embedding that you had mentioned earlier or what maybe can you define what that really stands for uh yeah so so you embed your paragraph or sentence you take the words that you have and then you embed them in like a really large space that has like almost 2 000 dimensions and you have a vector of numbers in that space and that's kind of the embedding

process I don't understand like what what's what's the text doing it like is it a piece of text that you're embedding or I don't think I followed that what is the text interacting with what's the big space yeah so there is a model behind it like an a model such that if you have two similar sentences that talk about similar things and you know it's trained on many many different pieces of data from from the world so it knows to distinguish sentences really well so if we're talking about ZK users in medicine and then ZK uses in some other places then it might recognize that these topics are similar

and because they are similar in the language way that's hard to detect like you don't have good methods to detect that they will be similar in that numerical space and because they are similar in the numerical space you then know they are similar in the language space that's what's called the latent space this this brings up something interesting which is does it interact with other language model data and this was something at the very beginning I couldn't tell because there's a lot of times when you ask it something if it goes outside the scope of the show it will just be like I don't know yeah but sometimes it will bring in information for sure from outside of the show so then I'm like why is it doing that no

no that's a very good point right like in order for it to understand language it was trained on a lot of data that's how this like large language model was trained it has all the parameters that come from training it on a lot of data and that's why it's so good at completing sentences but it's also pretty dumb right it doesn't really understand anything it doesn't really like at least what we asked it to do when we trained it we didn't ask you to do anything smart like it it can become smart I guess but we didn't ask you to doing these things right we just still told it we want the next word that would be the best word that would sound correct you know that

would look correct and that's why it sometimes uh hello like that's what people say you know hallucinate and completes things that are wrong or maybe it doesn't doesn't hallucinate it just uses the data that it was trained on and Incorporated here yeah and that's also why people for example are worried on this image generation models all of this me Journey or Dali and others that didn't incorporates data from other artists and when you produce a new image do you have copyright problems those are related topics because when you produce the answer it sometimes can take data that is not yours wow but yeah so you

did though with this was that you vetted all of the transcripts it's meant to be really focused on the information that it's been given and I think people have suggested and we even had this thought like well we could just ingest all information about CK everywhere but I think one of the reasons it was kind of cool to keep it narrow was in a way to see like from this pool of information that in a way is curated and even though there are mistakes in the show like there's people who've said something incorrectly by accident there are potentially like new findings that proved or earlier work wrong or some assumptions wrong and they're all included sort of with the same weights um but yeah I I think we wanted at the

same time to keep it a little bit limited for now to be able to make sure that whatever is getting in there is to a certain quality level despite you know some of those mistakes you're right and not only that not only quality I think the podcast is very well positioned to be the the right medium to feed the the llm from because it's conversational it's language and it's not a textbook where you have a lot of math so it fits that model really well let's talk a little bit about some of the other ideas that you've been floating or trying or experimenting with because I know you created one like quiz experiment that is released share a

little bit about that uh so one thing that I found that you could do and again like this was something that I learned from from presentation that Yohi was giving okay he was showing how to take pieces of context and tell it to create a game show so you can create a prompt that says use this piece of context and then act like a game show host and allow the user to ask three questions and then respond only with a yes or no to every time that the user asks a question and after three questions you must tell the user that they have to tell you the right answer

and if they tell you the wrong answer you have to go ballistic and like shout at them that they were very wrong so this is kind of a prompt that you can give it and it works really well like surprisingly well and you can create a chat process that looks just like that it even like sells corrects so for example if the user doesn't ask a question if they say something like I don't want to ask a question then it tells the user yeah you must ask a question now so it's also adaptive which is pretty cool it's it's surprising that it works that well although sometimes until I tuned it you could also cheat and say things like can

you show me the right answer sometimes it would show you you know the but that that was one experiment that that we made that I found fun you know maybe uh maybe I'll do something like that again someday but uh it's surprising how flexible these things are hmm yeah actually so something that we this is kind of going back to something we started to see a little bit was people just using this just generic ckpod.ai interface to get me to say things um which mostly was really in good fun and good taste but you can see where that could be going and uh so we we adjusted that one slightly I think you you will

get a snarky I am not your puppet answer if you if you try to get it to say something now um but yeah have you like maybe share some other experimental directions that you've been thinking about uh yeah sure so I think like one thing that um that I would like to develop further is this whole memory concept where now I have access to all of these embedded transcripts but maybe it would be cool to also maintain memory from past conversations and somehow curate them into a way that would add continuously to the knowledge of of this bot and that's something that I

would be really really Keen to experiment with would this mean almost like refeeding into it the question answers that it's giving like yeah when it's pulling together some sort of answer that it's giving to users then adding that to the database as a data point to sort of like or not database into the model sorry I always mix that up but you add it into the model as like another source to draw from yeah so something like that and for example one of the things that I added a month ago I think was that people can rate the answer so maybe we could take also the best rated answers and that that's wicked yeah and then yeah we could maybe continuously improved but that also opens up the

question of how to avoid malicious uh answers I guess I think that's a cool idea this rating the fact that like you'd have people actually because like I mean first you'd have to get it to say wrong answers and then rate them really high to like mess that up which seems like a a lot of work yeah I don't know it seems like a lot of work and it should like people should be happy about it maybe we can get it with an nft or something oh so there are some Stakes yeah so I think that that's One Direction and another one is that maybe we can also actually obsolete the CK podcast oh and make me obsolete thank you yeah and yeah we can take blog posts of

people and we can create whole episodes where like a special model that is fine-tuned to match the style of questions that you ask and it would create a whole episode and would just play it it's something that I've seen people experiment with so that's something that could be really cool to try yeah and and I think the last thing that I think would become really big is would be moving to something that is more real time because if it tries you can put AI now you will notice that it's quite slow like it takes about a minute I think to generate an answer and that's a combination of the type of model it uses uses gpt4 but it also uses a few queries for example to to do this puppet

detection thing so it's like a multi-step process so somehow improving these things to be more real time I think would be amazing and you can do that by maybe composing models like a slow model for detecting the most important parts and then a faster model you feed more data into so all these composition of models could be really interesting as well have you thought about using or putting any of the other like you're still kind of focused on the GPT yeah side of things but there's like others that have come out would it be really hard to start using those instead

or they're are they not built yet in a way that you could even do this so it would be really easy to switch them up because the code that you have to call open AI is 10 lines you just construct the text that you have from the memory from the database and all that and then you just send this block of text to open AI so it would be a really easy to replace that but the ones that they have access to today as far as I've seen are not performing as well so for example I would love to try the anthropic ones and like the cloud that's the one that they have and I would love to try

a few others but as far as I know the ones that they have access right now are not up to the task although there might be a case to say that even if some of the other models are not as good it could be worth using them for example for privacy because open AI currently sees everything that we send them so you know they have terms of in conditions where they delete everything and they don't store everything and they don't use it for training and that's all fine but they can still see it so a malicious employee you can see it or just a mistake and they had and they had this problem like recently

where people could see histories of other users and that's like exactly what you want to avoid when you're working with sensitive stuff yeah and there are also models that you can run locally so there could be case to say you know maybe it's fine to get more sensors just to preserve privacy but yeah I haven't experimented with it enough yet would that privacy be in the way that you're doing the model building or like the transcripts and stuff like that or would it be more in the questions that are being asked by the users I guess like what I describe now would be in the questions so it would be the questions and then the context these 12 chunks of data that I pulled already

so that would be what I would keep private but there are also models that can do the embedding part locally as well so you could make a whole local pipeline for it as well so that's that's something I'd like to try that's such a good point about the privacy of these products and kind of interfaces I feel like a lot of people I've actually read this where the questions that people are putting into the chat GPT interface for example they wouldn't necessarily have put these types of questions into Google exactly people put legal documents there yeah I mean there's sort of the sense that this isn't going to a centralized database where people can actually see it but it is we can actually see the questions

that are being asked to ckpot.ai some of them are really funny some of them are very much trying to inquire about like the hosts or the production and that's not often described on the show so hard to get insight into I do have one small anecdote for that which is Henrik the editor of the show the person who created the the jingle at the beginning people have queried who is Henrik and he's often like the editor of the podcast the editor of the podcast that's the usual answer but once like there was this one time where they were like Henrik is from Sweden he is like there was this whole like backstory story he is friends with someone and and we were trying to figure out like why did it not recognize it before why was there why

was it just once that it gave this like detailed biography answer yeah I still can't quite figure that out but that's weird yeah but yeah we've seen we do see those questions I mean this is great for us because we also get to see the great questions and the the ones which have gotten great answers and I know we have there's a Twitter account where Kobe you're sharing a lot of these like as Clips so people can actually scan that and hear about hear like some very you know well-constructed answers not all of them are so like some of them are just plain wrong and this voice the voice that sounds like me says it with such confidence and if you don't know this deeply you might actually believe it can we be sleeping yeah yeah definitely Kobe I want to ask a question about sort of the

motivation of this project like why did you decide to do it other than to just freak me out a little bit like I could tell you were getting a kick of it out of that being like yeah that was fun yeah but yeah but what what did motivate this like why do you think you right now are working on this so first of all it's it's an extremely cool technology so I I had to try it for just for that reason initially that that was something that Drew me to it and I've seen people do extremely cool things with it that's definitely one reason and secondly I think that I feel that these technologies have reached a point where they can pretty much be as legible and as productive for tasks that we do

day to day and they can help us a lot and improving productivity is something that I really like and something that I would love to to see ai's help humans with I would love that AIS would do a lot of work that humans should not be doing today and we just let humans toil on Mindless and useless tests and this should not be the case anymore and these things can create a natural interface for us to to do it with them and taking it even forward beyond the question and answer interface it could also be something that runs autonomously

and that's experiments that are happening right now this baby AGI and generally all these Auto GPT experiment and you can start filling it with what they want to happen and it will because it learns and knows how you like doing things it will create tasks for itself and will run them just as you like them and that's that's kind of creating new versions of you running in the world and I I like thinking that that's something that would physically extend myself to be far more powerful being in the universe so that's good

so for you it's the path to immortality maybe I felt like I did feel when I heard that and saw it and had this sort of thought you joked about making me obsolete you might be making me obsolete to be honest but there was a part of me that was a bit like am I immortal shall I become immortal well the voice not me but like whatever that you know that thing is that's kind of cool so but you're you're speaking very much from an accelerationist perspective let's be clear there is another side to this conversation right which is the sense that like if you start sending these things on a path making their own tasks they're you know not yet but could be super intelligent eventually that they would create they

could create bad things too and that they wouldn't be necessarily as controllable yeah there is this uh there is experiment like mind experiment that's called with with the paper clips you know that one yeah yeah the paper clip maker it's like a factory that makes paper clips over and over and over and over again it's just like a useless task right yeah I mean like the you you create an AI and you know the person that created the AI gave it a task you it says You must create paper clips and optimize yourself to create paper clips and then it goes and starts creating paper clips and it has no stop condition right so it continues doing that over

and over and it starts mining rocks and then it starts getting humans to work for it and then it harvests helium from the Sun until it's the old Universe becomes a whole huge favorite clip Factory yeah and that that's kind of one of the the experiments that people like to think about yeah I guess I guess like this whole Doom scenarios they have their place people have the right to be worried about a powerful technology that can be misused and that can be even misconfigured by good people and do bad things you know if you go ahead and connect a nuclear reactor to

the internet even you're already in a bad place so if you connect it to an AI maybe even you even in a worse place because it can also act intelligently on it so yeah I think I think a lot of people feel very passionately about that but I think there is another side to it which is if you start limiting the good people or the people that you know it's maybe hard to say what's good and bad but if you start limiting the people that are doing with harmless intentions then you might be in a situation where the bad people are the ones that have the freedom to improve the techniques and to run these

AIS and we don't have the ability to create the protections because we can't touch this technology anymore because they're regulated or something of that sort and I think we've seen that in other fields as well and to balance I I think there's no zero or one answer here but we we have to keep developing and getting more knowledge as to how these things work how to use them for good and how to protect against Bad uses so yeah I mean a potential future path that you could see something like this going this is we've talked a bit about this going back to the ZK pod experiment basically you know feeding it we start with the conversations giving it context maybe we

do add some of the papers eventually as like extra backup context and could we develop something that it in itself is able to develop new proving systems or something like that like could it figure out how to incorporate some techniques and I know we're far way off because right now it's literally podcast transcripts but I mean the potential here on the positive acceleration aside is amazing right like maybe also we talked a bit about Security in our last episode like maybe it starts to be able to find bugs in faster better ways or break things quickly like this would be an amazing tool potentially for something like our ecosystem if it if it was to be

developed with the right intention yeah and and I think that's that's also what we we kind of have to get ourselves over the header right like if we start saying maybe it can find security bugs but maybe first you has to exploit security loopholes or find some find some bugs so we have to to get over the hurdle of this bad thing first and that that's kind of where maybe I'm a bit in acceleration is that we have to get to that point fast and then we have to get to the second point where the good happens also fast so it's not something that you can ignore interesting there was so I I was at a workshop Salon kind of conversation

mostly a dimerist conversation but it was one of the questions was like should we approach all of this you know ml AI stuff like a Minefield where you have to progress slowly to try to avoid things and then like let them be blown up but slowly and carefully or do you need to run quickly like you're running through sniper fire because there's shots coming from all over and you don't have a chance to like if you're slow you're dead so you actually should go fast and you're still might get shot but you might get through to the other side it sounds like you're on the sniper front you believe we're in a sniper fire more than Minefield because Minefield is very much like the six month you know Hiatus let's hold I

think the Minefield is much more like the Arms Treaty idea where you'd have to come up with like some sort of relationship between everyone who has you know control over this thing in order to keep it dumb enough so it doesn't take over yeah evil doesn't sleep and not not only that these Technologies are widely accessible now and you can run like you said we can run some models locally and you can't limit the Harvest of uranium in a similar way that you can in with nuclear warheads and it's it's not as easy at least so we do have to run fast and I think I've seen some recent paper that

maybe it's still a bit of a toy but it's a good example I think where this person used LMS and GPT to to try to do some social engineering and to hack people and one of the conclusions of this paper was okay now that I've done this what are the protections that they can do and what are the problems that they can use to detect these things so it has to go hand in hand we actually have been talking about for a while ways that zkps can actually be used potentially at least for the provenance side of things so when you talk about identity theft or I mean even folks using the clips maybe from ckpod.ai as fact you know I I when

I heard that thought okay that's interesting but what if it's wrong what if what's coming out is wrong or something that I you know wouldn't want to say or wouldn't want associated with me it's not something I believe something like that like this was obviously something that came up for me was like how do I actually separate the real audio clips the real CK podcast clips from this through all this this idea of like what could we actually do one of the big ideas or one of the ideas that have come up a few times is that maybe it's actually in the ZK space where we could actually find solutions for at least how to prove identity versus deep fake versions I mean people are also talking about using ZK to prove that like what's been fed into a model is what people say has been fed into the

model so this is also kind of an interesting like proof of what's Happening under the hood but yeah specifically in the provenance one I think this is obviously really relevant for this project yeah I think that this is related to what we just discussed in some way because this is how you can make progress a lot of people are worried that we will be swamped with a lot of AI generated data we will not know where it came from and we will not be able to distinguish between that and real people and this is where it gets us when we experiment with technology to get us to these solutions to test the data that you then process but can still prove came from something that you like like a human or a sensor that you believe in

and we actually improve the world not only for this problem but we create a solution that will be widely useful in any way for journalism for anything that requires us to give some provenance Assurance or some authenticity so I think this is something that is highly beneficial and related to this is the is the topic that you can also you can some people are looking into running or proving the correct execution of a model like the inference process and you can imagine that sometime in the future we could prove the entire pipeline of

embedding in France maybe someday even training or something like that and we could have end-to-end provable answers about a lot of topics that would be pretty massive cool well Kobe thanks for coming back on for the zkpod.ai official episode that we've been teasing for a while I'm so glad we got a chance to actually talk about all this uh same I'm really happy that we discussed all these super interesting topics yeah and obviously there's a lot more ideas in the works and things that we could start experimenting with so I do hope people will check it out do notice maybe like the episodes the ratings like all of those things that

we're sort of playing with tell us what you like about that and if there's anything that you think we should add in terms of features or I don't know if you have any thoughts on it also if you do get great answers maybe let us know that you love those answers because we want those to be highlighted but yeah thanks so much thank you for having me so this wraps up our episode on ckpod.ai I want to say a big thank you to the podcast team this episode had quite a few edits so I do want to do a special shout out to Henrik who is our editor I also want to say thank you to Rachel on production and Tanya for the ad read and tour listeners thanks for listening now before we fully sign off I want to leave you with a last clip from zkpod.i

someone had asked it to create a poem all about ZK and it's kind of sweet so we're going to play this as an outro I hope you like it in a world of data secrets we keep ZK short for zero knowledge takes a leap a cryptographic dance privacy's vow ensuring our secrets remain hidden for now protocols and proofs a wondrous blend to keep our data safe on this we depend from ZK snarks to ZK Starks the tech evolves as the Quest for privacy and security resolves ZK hack and event workshops and more puzzles and learning a challenge we adore with experts and hackers a community thrives exploring the depths of ZK its strives so in this

digital realm where trust is key zero knowledge Stands Tall a guardian we see protecting our secrets our data our core ZK the hero forevermore