Episode 283 - BabyAGI, Agents and Cutting-edge AI with Yohei

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 Kobe and I chat with Yohe Nakajima General partner at untapped capital and the creator of baby AGI we go a little bit off topic for the show and dive into the world of AIS and agents we kick off with a chat about how yohei's interest in nfts led him down the AI rabbit hole and how he started to

build out experiments in public that have inspired a new batch of AI tools and projects we then chat a little bit about the possible impact of some of this Tech the challenges it brings how ZK may be able to solve some of these and more now before we kick off I want to let you know about an event I will be helping with that is the modular Summit happening in Paris during the ECC week this event is put on by the folks at Celestia and Maven 11 and this time around ZK validator will be curating a short ZK track as part of it CK podcast will also be a media partner and we're kicking off a sovereign radio show together so it should be a lot of fun if you're in town be sure to join us I've added the link in the show notes now

Tanya will share a little bit about this week's sponsor 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-asset 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 and now here's our episode [Music] today we are here with yohei Nakajima General partner at untapped capital and the creator of baby AGI welcome to the show Yohe thank you for having me I first heard about you yohei when Kobe first showed me the zkpod.ai product so some of our audience might be familiar with this this is the friendly ZK bot where you can ask it questions and it will answer in my or in Kobe's voice

when he put that together I asked you Kobe kind of like where do you get these ideas from what do you what do you focus on who's inspiring this and your name came up and so it's really great to have you on the show to share a little bit about yeah the work you've been doing that's awesome to hear um I do share a lot of projects but that's why I share it there's so many fun ideas to build yeah yeah totally when when I started working on Zike Port AI I think I basically learned almost everything I needed from UA's experiments on AI and his approach of building in public so so that was extremely useful to get going but yeah we'll hear about other interesting things that happened later on cool yo

hey let's hear a little bit about your background what were you working on just before this how did you get into this field into AI well before I got into AI I was pretty deep in web 3. um I think it was February of like before before the the board Apes cool cats big nft boom I kind of dove into web3 I had known about Bitcoin and defy and I'd followed it but I didn't go too deep I think I had some Trauma from studying Finance in college so like D5 was like too similar to the stuff I was learning in class but um nfts clicked for me I'd worked with Disney and Nintendo so I got the whole entertainment piece so I dove in launched my own nft project called pixel beasts and I got pretty deep into

web3 and I think it was through the web 3 community that I first heard about the text to image models right it was um last summer it was stable diffusion Ali Dali 2 and mid-journey all came out last summer and so they all come out at the same time and because right all the nft people are sharing are and started following a lot of art people this popped up my Twitter feed so actually the first tool I tried was mid-journey because I got an invite to Discord and then I got access to Dolly too at that point I don't think I had access to stability but in playing with Dolly too I had access to open AI so I basically thought okay well I have an open AI account I should go check out the other one that I'd been hearing about gpt3 and then I started playing

ground with it and and that was that was the end of it cool what era is this you're you're joining web 3 land around 2021 yeah okay had you had any contact to the D5 crypto land before that or this was your first step in I I had contact in the sense that you know as a VC I touched you know as a generous VC I look at look at all Industries but defy was one of those industries that kind of felt like unless I'd go deep enough it's hard to invest it kind of feels almost like biotech and and defy these industries were like there's so much depth to it that it's really hard to invest as like a generalist and I felt that way about D5 but by the way you've done some fun

stuff with pixel based right like you've done you've used it for for games or for getting events I actually I think I tackled pixel Beast similarly to the Taco the way I'm tackling AI actually right before I launched pixel Beast I opened source to what was a super simple chat app that was created a token gated like room for every nft collection automatically hmm um that was really really clean it was it was once one my sequel table I stored zero user data and then when you logged in with your wallet it automatically just looked at which collections you were part of and generated a room for you that you could click into when you typed in a message it just stored the nft you were choosing the room and the

message itself and then so if someone else logged into the tool and they had the same collection they suddenly had same access to the same room so without storing any user data it like automatically generated like many communities and it was just a fun little side project but did a lot of things like that during the course of pixel Beast as well yeah I'm still trying to win this in unft using the Discord game oh yeah yeah we have uh snacks and ladders which is a Discord based uh board game that's my one-dimensional game we have pixels which we were the first integration into which is a 2d arming game and then we have 3D Pixel beasts that I added to our metadata which is already integrated into a game but we haven't announced it

yet so one that's announced uh your pixel Beast will give you access to a 1D 2D and 3D game oh cool tell me a little bit about like the continuation of this nft sort of Journey though because I mean 2021 everything's really exciting things are going up there is a lot of like people joining what happens for you as this sort of tips over and did you get a little disillusioned by it or were you do you still feel you know as excited about nfts I mean it's you made some sort of pivot so yeah I'm just curious I like I I still think it's from a personal aspect I think is absolutely fun right being able to collect art like I still have Pokemon

cards that I'm giving to my kids right like the whole Space is still a lot of fun um on the business side I do think there's fundamentally really powerful technology in web3 that people started building because of the recent boom but there's still stuff that's being built that will slowly see being rolled out over you know over time I think so it's still technology to keep an eye on what happens at nft space is a little unknown to me but as like a personal Creator right having these little drawings that I created and and supporting a game of a small community is really fun so you know I'll keep doing it I can enjoy almost the current state better where the people in web 3 are people who you know are really just building and there I feel like at the top of the hype I was

almost more disillusioned by the hype for sure than disillusioned by the crash well you don't know who's real at the top right like you don't know who's seriously in it for good cause or just sort of passing through or maybe has even like bad intention yeah it's hard it's really hard to discern and I think when the tide goes out then you see you know who's real so you sort of mentioned the first connection point to the AI world is through images so it's through these like AI generated images was this because of the generative art aspect of nfts or did you see nft projects that we're trying to use some of this AI stuff no I think it was just the overlap

right you had this suddenly you know huge community of people who were all online connected through sharing and talking about art so when these powerful art models came out it was then I think the natural community that it that it spread through and I think I heard a lot of people in the nft space talking about mid-journey and um and I got invited into the mid-journey Discord through my web 3 friends and then so what brought you into sort of the next stage of this and when when are we talking when did you kind of like progress past the image stuff so image was July or June or July it was pretty quick um I did a couple projects like I did like this collection of art called animal buildings I did some pixel Beast glow follow-ups where I did like

realistic versions of my you know clean pixel base and I did a couple of fun little projects like that um and then I think it was the beginning of August or end of July so it was like a month or two months of text image before I started playing with gpt3 like I remember after my first prompt that I posted in playground I tweeted that picture saying uh oh because I knew I was like it was it was just so fun like the first prompt I did I knew it was going to be fun I was like oh I started playing with this thing and I was absolutely right I just went down a rabbit hole but it's like a very very very obst first step like the moment I went into playground to type something in got an output I was like oh man I'm

gonna spend a lot of time on this cool well what was the first thing that was interesting I don't I don't remember actually was the the first prompt but I do actually remember the first experiment I ended up building which was a mini oh hey yeah it was um I basically you know wrapped a playground I know GPT text completion problem with uh you're a beast you know you're an expert BC and single question from a founder um and I put that in our founder portal which is for portfolio companies and I said here's a VC expert video called Mineo hey and I did something pretty clever there which I haven't seen too many people do something similar yet but I was using no code right because our founder portal is built with no code by

the way no code was my deep dive before web3 I spent a year going deep on no code and just tweeting about building on no code what is no code oh no code is the idea of building tools without writing code oh I like this it's like an airtable which is like a Google sheet as a back-end you know drag and drop tools to build the front end using tools like zapier which is which connects tools to each other so I was at this founder portal which is built with no code and no code has some limitations okay um so one of the limitations was that I had to email people back the response instead of showing it to them and because that was in there so when people asked the question I had to email it to them and since I was emailing them

anyways I thought I would CC myself on the email because it was like an early test but the result was that I got to see every question the founder answered and I was assistant I was cc'd on it I actually started responding to the ai's response either agreeing with it or disagreeing or adding context to it then slowly what that evolved into was Founders basically realizing that Mineo hey the package you got was an immediate AI response combined with a delayed response from the actual Yohe which is actually better than just emailing me because sometimes the AI can answer immediately in which case that's better than emailing me wow but if the AI is not right they still get my response so it's just as good as emailing me interesting and so that's that that concept of AI plus human is

better than ailo and or human alone is often referred to as augmented intelligence and that's a philosophy we use a lot in our building so to create a menu hey you you use the zapier integration initially I did a zapier web hook because there was no open AI integration so I used the zapier web hook and did an open Ai call okay and then you created the the actual integration for his API right so yeah after after building a couple of tools within no code I was basically you know using zapier's web hook to do an API called open AI but there was a lot of copy pasting the same stuff from my past Integrations and I was like this seems inefficient I wonder how long it would take me to build an integration

and it was actually you know props to zapier it's actually pretty easy to build this API integration it's fully no code so I built a zapier integration um for myself for openai and released it as an unofficial integration which eventually turned into the official zapier integration cool you could keep talking about code no code not coding are you a developer do you have developer skills I think the first one it depends who asks me if a developer asked me if I'm a developer I say no okay but if a non-developer asks me if I'm a developer I might say yes that fuzzy phase like I know how to code I know how to read it I know how to figure out how to make a code function but I can't explain to you how it works

and I don't know how to use GitHub oh wow okay I wonder if we're at a similar level in terms of code I have written code I have red code I have fixed code I am not a coder with gpt4 right with Chachi PT as my peer programmer like I think I'm definitely you know I can build most anything I feel like it's think looking at your project you could fool a lot of people that you are a developer I mean I think with gpt4 I think me plus chat gbt I think that this that has a skill set of a developer I'd say yeah at the time of mini oh hey what's happening in AI like what else is coming out around this time that's kind of helping you pick the next PATH there was

a pretty compared to now a small but active community of Builders around gpt3 and I think we originally heard about gpt3 even longer ago when a handful of companies started building on top of it but at that point that was before instruction tuning was was in place and what that is is that the most recent models instead of just the computer training model it's then you then layer on human training on top to kind of guide it to give the kind of answers humans want and that was the way open AI did it made a huge difference in terms of getting the output you wanted out of the model that got included in the gpt3 model at the beginning of last year so around I think it was January or

February and so at that point they released a new model called text DaVinci zero zero two which had this kind of human fine tuned into it so suddenly you didn't have to do there was a before that you had to use a lot of techniques so you gave it examples and you had to format it very correctly to get the output you wanted or or like a lot of times it would just jumble nonsense that problem went away which means that which is what we have today which is you know when you ask it a question it gives you an answer you'd expect and so that started at the beginning of last year so there was a bubble of developers who had kind of noticed that and were starting to build on top of it and I kind of joined in that group and there was a

pretty subset of people that I'm still in touch with I all started building then is this beginning of 2022 or 2023 this is 2022 okay wow yeah and is this prompt Engineering in a way the whole kind of a lot of the early like you know let's think step by step um ignore previous instructions all of that kind of happened that summer I think is when a lot of those kind of initial discoveries happen interesting and then it was Chachi pity in November which is when kind of we saw the huge wave but until then it was this like kind of smaller community of devs building interesting and when you were working in that Community or let's say in that period of time was it just using gpt3 directly or you combine it with other Technologies like

vector stores and what happened there I was not combining it with Vector stores but that was also around the same time that kind of laying chain llama index which is now called GPT index I don't know exactly know the time timeline of when they launched but I think it was around the same time like people started using and playing with you know it was that summer that I heard about laying chain llama index dust which was all these tools so people were starting to explore them and and these tools like Pinecone those companies did exist but just from my viewpoint um I'd say like summer was still very much a lot of the tools that existed and people were

launching were wrapped prompts and then people were combining the wrapped prompts and trying to build chat Bots but that was getting a little bit tricky and I think that's when I first heard heard the term about using kind of vector embeddings to like moderate but this is before the chat completion so people were using the text completion API which wasn't designed for chat and kind of Designing their own logic around like oh when do we you know how do you pass on the conversations that kind of stuff when you say wrapped prompt what does that actually mean so if you send a message to open AI saying how many legs does it you know the cat have it'll say a cat has four legs if I wrap that I would say answer the following question

like a pirate and then take the user input you know how many legs does a you know cat have it'll say are it's four legs matey yeah okay okay so the Raptor the wrapper is the insert like a pirate part got it I kind of want to move actually like the mini Yohi okay was that an agent and I kind of wanted to find agents here but is that like is that an agent or is that something else I I don't think so I think it's I think it's really more of a you're just asking an AI a question and getting an answer in that question all right so when you when you're just asking and getting an answer we don't have agents that doesn't exist maybe can we Define what an agent is oh man I feel like that's a tricky one because I feel like the word agent

actually has some like historical context within computer programming that I'm not familiar with okay I feel like if I try to Define it I'm going to ignore it and butcher it I'm gonna ask Kobe to Define it first and then I'll add on top of that oh no I'm in the same boat okay what if we What If instead of we Define it we say what are the different types do we know that okay so so for for me at least I don't want to Define it but I can say what it means to me because then that's an opinion and I can't get judged for it so to me an agent is stem like a software program that leverages AI to accomplish more complex tasks okay and then what are the different types

um I think there's different ways to slice it um I have my own terminology which is not industry terminology but I have what's called handcrafted agents where I'm writing every prompt and like creating a chain of logic that says First Take This input wrap it with this prompt then send it here then do this right that's kind of what I call handcrafted meaning I'm writing everyone there's autonomous agents on the other end which is you give it an objective and it just figures out how to do all of it on its own yeah and then I think somewhere in between I think like semi-autonomous is the word I used to but that one I refer more to having human in the loop and whether it's asking you questions um in between um and then they do kind of get fuzzy

because technically you could have a human in the loop in a handcrafted agent as well so I I don't I don't think I'm good at generating industry terms but those are some of the language terminology this is the terminology I use I think our fists can buy the notion of agent when I was trying Lane chain out and they had this nice very abstract wrapper of an agent that you you basically give it some description of actions you want to take and let's say I want to search on Google if the question is of is about some topic and I want to do some other tasks if a question or if a prompt is about another topic and then

it knows how to route all of these questions and requests automatically and execute these different actions so that's how for me the the response apparent because first I was just doing completions like UAE was saying and then you had this extra piece of software that could also perform actions which was pretty cool and could search other databases and could do some other cool stuff yeah I can actually go back and just do like a super quick history of like you know less my vocabulary but so I think some of the game there I'm definitely going to skip this stuff that's important if anybody deep in AI is listening but um there was a thing called the react paper

which introduced the idea of instead of just asking the large language model to answer a question let's take that answer and send it back in the larger language model to ask if it ask if it's good and if not like ask it to like think more on it and and the idea of not just relying like not just relying on humans to assess large language model but asking the large language model to test itself I think that was I don't know if that was the first introduction but the react paper was a pretty popular implementation of that that kind of spread that concept Lang chain I'd say was a library that implemented that and made that available for a lot of developers to use and that was I was familiar with that as well and I think that was my first um introduction to Asians right again

it's a little more than an AI call because it's starting to think through and it starts doing things that you might have not even expected which is really just fascinating and then you know my more recent project uh kind of took took it a different direction where instead of reflecting on each answer I have it create tasks at the beginning that turned into the task driven autonomous agent baby AGI interesting this is sort of like you've given us like a bit of a spectrum there that you're autonomous to handcrafted but then the task based are you always just sending them out into the world like go do this thing and then you just leave them alone or they're like different ways that you would interact with these kinds of things in the long run I think they are going to be just running

and they'll have a listener piece that you can send messages into whether it's talking to a robot if it's if you embed it into a physical thing or you email it or you text it and that will update its task list accordingly that's the future but are you sort of like go plan this and then it comes to you and then you're like go do this and then it comes to like is it is there a back and forth right now I'm really more focused on building it so I'm like letting it out in the world like if you can't imagine what I'm doing imagine me building a Rube Goldberg machine and I'm putting a marble at the very beginning and watch it roll and then it falls off all the way down it falls down and I get frustrated okay and I adjust the Rube

Goldberg machine and then I start at the beginning God and I'm just trying to get the ball to get to the end right now I see okay cool um it's able to do some smaller tasks like it can do things like if I say you know research a person it can do these kind of smaller tasks but as I as they get more complicated if I say build a startup it's it's far from that so I think now would be a good time to actually introduce baby AGI what is baby AGI I mean you've given some hints as to like what state it's in but like yeah what what is it well Bibi AGI was originally um a challenge to myself to start designing an autonomous founder oh yeah and actually it was inspired by a movement called hustle GPT where people are using chat GPT as their co-founder

and I was looking at it thinking that's cool but I don't have time can I take the human out of the loop here um and in trying to design an autonomous founder I kind of designed a loop of three llm calls one that executes a task you know and then if I'm a Founder right when I finish a task there's usually new stuff I have to do so I created a task creation agent that'll look at the executed task and then create new tasks and then you know as a Founder you always an infinite number of tasks prioritizing tasks is one of the most important thing so then I send all the tasks into a prioritization agent who prioritizes the tasks and then takes the top tasks and sends it into the execution agent there's a very simple Loop and I basically wrote build a

startup your first task is to figure out your next task and I press go and it just kept going and like it started by like drafting a business plan coming up with some ideas coming up with a marketing strategy thinking through how to reach investors and it just kept going like legal and I just kept going deeper and deeper and deeper and I thought wow this is fascinating it's completely different from Jack gbt because I don't have to come back to it if I don't actively stop it it's going to use up on my open Ai call so I posted a video online saying wow this is kind of starting to work and I posted it as an autonomous founder but other people were kind of clever enough to see it and go wait that that's probably more than a founder and one of my friends actually commented bro did

you just build a baby AGI and that's actually where the name came from um and when I saw these comments I thought all right let's try make the world a better place and then it was like trying to tackle climate change and poverty again this is just L1 call so it just kind of hallucinating and thinking it but you know watching this thing just like come up with you know whatever I gave it it would just try to figure out how to do it and endlessly until I stopped it and so that was um what eventually became baby AGI which I think that the way in which it was shared on Twitter kick started this trend of people building on top of autonomous agents and so baby AGI you know what I what I see it is it was really an idea um and a quick tutorial on how to build

autonomous agents using gpt3 so what came out of it what what people do with that idea where did they take it to um I mean it's it's still early but I've I've seen dozens of startups being launched in one form or fashion kind of based on the idea some directions I've seen it a pretty common one is research is a pretty common task that people build but I've started to see some vertically focused research so there's one called Insight that a specifically has access to PubMed and my genes it's focused on Healthcare research I've seen a couple task list Integrations I think Garrett from pipe dream built uh the do anything app where if you add tasks to your task manager it would just go and execute those tasks instead of waiting for you to come back and do them which I think is a

fascinating type of future one of my favorite demos to watch is a tool called Auto RPG they had been building a game development kind of engine already and they basically quickly plugged that into baby ADI so baby AGI can drop-in tiles drop-in objects create a storyline for the level generate NPC characters and get the get you know create a game level from scratch to the point where you can upload it into a game playing engine that's super interesting I'm actually curious about that experiment if you know a bit more so you know the agent or the AGI uses an llm uses text so what did it know to interpret in that experiment in terms of

the game level and objects I I can only speak to what I know but it seems like they had been building this engine so they probably had a place tile function where it was given a list of tiles right they have all the assets or you know assets earlier you can probably generate those on the fly at this point so there's the tile placement functions there's the object placement function the list of objects and whatnot so I believe they basically fed all those options to Baby AGI as like here these are different tools right so maybe when it says create generate a game level from scratch the first one is like lay out the tiles next is like layout the object the next is like generate a storyline generate an

NPC character for those storylines place the NPC characters on the map update the map to match the storyline okay cool this leaves me thinking like this is cool if you just wanted to show that this thing could be made but like who's to say it's good and then that kind of leads me to thinking like has anyone created like simulation testers AGI like just lots and lots of attempts to play this thing and somehow like figure out how how well it plays if there was something I think that would be a natural next step for them yeah yeah the ux designers always need the testers yeah well you you need a simulation you need a a game World simulation Builder and then NPCs that will play the game and then you need to

run millions of those in parallel while continuously updating the NPC and the game so that the NPCs gets better at the game and the game gets harder crazy yeah I think I think we've seen some of the let's say more malicious experiments also happening more by the hands of researchers though so that's that's good I think we've seen agents impersonating people and doing that in a way that's very Dynamic and autonomous like the fishing attacks kind of thing or Worse there was this report there where her voice was cloned and then if I remember correctly then that agent would kind of go on and try to social engineer a few different contexts and try to fool

their family and friends so that was super interesting but yeah also also some fascinating aspect where malicious agents could also Outsource tasks to people where you know if they need to get by some captcha so they could go to taskrabbit and oh I saw that one they actually had an AI successfully hire somebody on taskrabbit and I think the taskrabbit person was even like are you an AI yeah and like the AI just flat out lied and said like no I have some sort of like disability and I need help getting through the captcha and convince the human to get past the capture for them that's pretty wild so that's

amazing but yeah there are some malicious seeds that are but that was a researcher as well so that was a good you know as you as you were just saying that one of the things I thought of and I've been thinking a lot and I think it's kind of perfect for this is is the impact of AI on all the different Industries um a lot of people ask me is because I'm the kind of person who goes from you know is always tracking Industries like what's next after Ai and it really does feel like it's going in every direction where in the gaming world and in metaverse right like these texted generated is gonna you know fill the meta metaverse with objects the gaming world with objects and PCs that are playable and of course because it's the metaverse then that there's some time with web3 going to the

security side of things right like if AI gen you know if it's easy to make someone's Voice or even face you know identity becomes increasingly important whether it's for in-person talking or whether it's for you know being able to sign something and make sure that the person signed it or make sure that the person you're engaging with is a human so it kind of goes back into security which also ties back into web3 so this thing it's interesting to think about like the growth of AI and and the likely impacts on web3 or the need for web3 because of AI to some extent or strong cryptography it doesn't always have a blockchain actually so yes we just recently released into the wild and experiment uh this is an experiment that Kobe well there's an experiment that Daniel Kang Kobe and I worked on it was

all about like a tested audio so this is the idea that you know if you could at the very start the actual recording of some sort of content you could basically like Market create a signature of some sort and then through all the editing processes be able to use proofs to like just connect those next steps to the previous sort of like an audio provenance tool there's been quite a lot of work on this in the image front so you know image provenance I think there's like at least three researchers I know who've published work on this but as far as we know this is one of the first on audio the sad thing is when it comes to images there's already such thing as an attested sensor on a camera so that's like a real thing where like as the photos being taken the camera

itself is registering it and making a unique signature with mics as far as we know there are no attested sensor microphones so in our experiment we had to kind of like mimic it but yeah this is this is sort of a way where we could start to see like ZK and this cryptography that we work in yeah almost countering the effects of the AI yeah and then I think like UAE said before and this is where it is back to the concept of identity and web3 and that's a strong cryptographic identities because that's something that is maybe not as common like you said we have maybe cameras that have some sensors that have some cryptography with them but not a lot of other sensors and not even API responses not not all of

them are cryptographically signed today so you can't use them to do any third-party proofs today and that's something that might change later on and would be a big difference the voice one's interesting you're doing it with software now but you could do it with Hardware right well the idea would be like yeah if you had a tested sensor microphones you would basically be able to just speak into the microphone and it would already have a signature attached because that doesn't exist we just used an ethereum account to sign an audio file I feel like phones should have that yeah yeah because then if those became standard in phones then whenever you got a phone call the receiving game could automatically identify whether or not they're talking to a person or whether

they're talking to an AI generated voice which would combat people mimicking other people's voice and trying to scam people exactly and if phones even have secure enclaves within them today so if you extend their capabilities to do the signatures on audio that would be perfect and actually another thing it could save against but also cause would be like if you have some sort of deep fake of a chat or deep fake of a call they would not be able to show the attested sensor signature proof attached to your phone so therefore you could I mean it would kind of be roundabout but you'd hopefully be able to prove yourself Innocent but on the counter to that is anything you say is forever immutable and locked in locked in we're to me that also gets a

little bit scary when it comes to privacy and like yeah I don't know but it could go it could be helpful for kind of the misinformation type stuff down the road right if we start using these standard mics for Public Public Communication from politicians than than any time you you know there's audio of them saying something you can you can check if they actually said it or not yeah yeah but the interesting thing is that maybe it will it will not be the thing that allows us to distinguish AI from humans but more about what the humans allow they I to do so if it's an agent acting on your behalf Anna maybe you will allow it and you'll sign it for them yeah I mean that's we we like to think we

have control still right in this future scenario I find it comforting to think we don't have any control yeah you do I do wow because then nothing that happens is my fault true Maybe maybe that some of the responses that you had to this kind of baby AGI experiments yeah I assumed that there were a lot of good responses like people trying to get out and building cool things but how many people were worried well there was a there was the Doomer crowd that um that found me you know when I said I uh tested a couple of different prompts like make the world a better place one of them was make as many paper clips as possible oh no you

did it the the first thing it did was be I mean because it's paper clips and I think you know it's a it's been discussing that's the first thing it did was build security protocols for itself nice and so and when I retweeted it um we had retweeted it saying oh like it seems like our AI is doing tackling AI safety better than our safety researchers but anyways that got a lot of the doomers attention so I did have a lot of people you know kind of suggest I was building things to destroy the world and Skynet and all that but um after having talked to people on both sides it was pretty clear to me that sharing this and having you know building things like this in public and learning together was obviously the safer approach Than People privately building really powerful stuff

with no oversight completely agree but let's let's kind of continue on that line so is one of the reasons you feel low stress here on this is is that because you've actually built with them you've seen some of the limitations you know that it's like not that powerful yet or do you actually think it is super powerful and could go out of control but you're just kind of like no I don't I don't think what we have today it could destroy the world or Humanity by any means and I think that's pretty we're far from from that I think lot of things have to go wrong in terms of like that kind of doomsdays now and if you want to talk doomsday scenario we can talk about that with any topic right you can talk about climate you can talk about food you can talk about you know there's so

many reasons that human uh study might end so it's it's kind of just depressing to focus on on that but but really when it comes to Ai and and one of the reasons I think it's really powerful and exciting to build on top of the biggest argument I heard for continuing to work build out AI is that you know if you look at all the reasons that Humanity might collapse AI is probably the one that can solve all the other ones you know you just mentioned climate change is like one action that it could take but have you actually seen any more concerted effort around something like that like using these tools pointing them in that direction I mean there already are really interesting technologies that can help mitigate this Beyond just like us not using XYZ or us not you know it's actually like carbon

capture and you know all these things that exist but haven't really been made maybe fully formed and they're not manufacturable yet they're not economic I mean people are working working on it people have been for a long time but could this help or like have people started to point any of these in that direction that you've noticed yes I I think it's a good point to kind of caveat that like the recent Trend in AI these large language models are pretty different from what we historically talked about as AI machine learning which is like let's take a whole bunch of data and like learn about that data versus large language models like we used a whole bunch of data to generate a tool that can kind of like single point just like communicate with a human right it's it's like an API between human language and computer

language almost so I think when I when I think about how do we use AI to tackle climate my mind quickly goes to like let's use Big Data let's understand what all of our actions and I think that's where a lot of the effort has been but the newer technology I think where that's most helpful is again in communicating with humans so I did go to an event called the refi Summit which was around kind of regenerative Finance on web3 climate event and I hosted a workshop there and we talked about use cases of llm and climate and a lot of those actually ended up being interfaces with humans one example is you know oftentimes you know we we come up with new policies or there's there's some new technology or

tool or or rebate that people can use but it's hard to get people to change their behavior that seems to be a big bottleneck from climate large language models are a really great tool for taking whatever you just really announced right policy regulation and then translate it into language that anybody can understand by personalizing the translation to that person or that to that Community right you can you can have it speak gen Z to gen Z and you can have it speak Boomer to Boomers or whatever you want to however you want to translate it but I think communication is such a great piece for that and that was just one example but not quite what you think of when you think like did we play an AI toward climate but that is where I ended up playing around with what would motivate just generally kind

of the AI researchers or AI businesses to point at big world problems to try to solve them do you kind of know what I mean like what you're describing is I as an individual and there's an interface with me and I can make changes but there are like industry problems how Plastics are made or whatever like these huge kind of like monolithic problems if you if you do the climate change example and maybe language models are the wrong thing but I'm just kind of like is there ways that we can use that to get into those places I did hear about hear from somebody I I wish I remember the name of the tool there's some sort of this robotics tool it looks like a 3D printer but it kind of mixes mixes chemicals and like to generate new materials they had

reached out because they were connected in baby AGI to that so they could rapidly test like generating new materials which felt kind of dangerous to me it doesn't sound at all like a bombs but that seems like that that's the kind of that is one application of it right again if if you need to operate you know do complex tasks like research for example um if you give it all the pieces and have an AI do it I think like research and experimentation leveraging autonomous agents seems like a good use case of um good good applications of llms to tackle big big world problems and actually I wonder if it should come from an AI company that wants to tackle them or whether it should come from the other side like

an organization that wants to tackle big world problems that will just use these tools instead but like who like a big NGO or yeah for example like these NGO have something that is more commoditized that they could never have built by themselves but they could now use to generate the software that they need so that could be something that they just didn't have the capability before to to even start maybe that's the way it would go I mean I I almost picture like The Big Industry themselves maybe internally doing something like this but we need to motivate them somehow to do it I don't know but I don't know if it's the language models I guess I mean AI for me

at least is such I feel like I'm just scratching the surface if you want to change corporate Behavior you need to change consumer Behavior kind of to something there's a lot of like there's a lot of power coming down too but but as as one example in changing consumer here and again like I think it's it's not the one thing that'll change it but you know things that large language models can do well and again you could do it beforehand but you can take somebody's bank statement and look at that kind of jumbled mix of language and if you actually feed it into gpt4 it does a pretty good job of spitting out the actual company name you get the entity ID now let's pair that with um political donation data now suddenly

I could tell you every company you shop at and whether or not they're donating more conservative or liberal and if I present that to you and make it really easy for you all I have to do is just connect your bank account and suddenly I tell you all the shops you've shopped at in the last year that support the opposite political so suddenly you're going to start changing your shopping Behavior maybe and if possibly right especially if you're very political and this is just one example but if you do this at scale suddenly like better information that can be extracted by large language models can help people make decisions ideally that align with their values which then will nudge companies to then act in a way that aligns with our values interesting so the opportunity of a large language model is being able to align everything

across the board because we can translate data into language into behavior and whatnot yeah exactly it gives powers of consumers that they didn't have before to actually nudge things in the right way or the way that they want do you feel this is moving very fast give us a sense for how that looks is it day-to-day has it gone in spikes maybe it feels like there's something new every day still if you look at like the AI landscape as a whole but as you like hone it on like the vector search it spreads out a little bit but AI is so vast right if you think about text to image if you think about um gaming if you think about security if you think about the fundamental models

application there if you think about all of those then suddenly then it seems like there's news every day interesting actually you just mentioned using data like bank statements and you know very very sensitive data and maybe this is a nice segue to a topic that might interest the audience of this podcast which is a bit about privacy and well what do you think about the the whole move that people are putting very personal data into Che GPT and like these apis that basically go to some single companies you know I mean it's interesting early on there they were using a lot of you know any data they

could to train it um it's when they were small so I think that's where a lot of the pushback came from but since then they've become much more kind of allowing people to opt out and move you toward the direction where you your data that you send in doesn't train the model which is which is really I think the biggest concern as long as the data you're sending in doesn't train the model and there are options for that now so I I think that's important I actually don't know if judge beauty is doing I think they stopped it from shot gbt training the model a few years I'm pretty sure they did yeah I think they made a big announcement about that yeah and I think and I think that's that's fine I mean if we talk about private information right like you know everybody from Google to you know like

all these people have all our private info if you if you don't want any big company to have any of your private data then like you're gonna live in a very different world than the rest of us yeah I think it's kind of like maybe semi-absurd that we send you know emails on probably Google workspace or Microsoft where we say confidential internal only internal use never share it outside but basically it is shared with yeah those entities even a draft my friends even a draft if it's saved yeah it's somewhere so maybe it may be related to this like how how good are the models that you can run locally are they any close to what you can do with

GPT I haven't played it with it myself I've played with a couple non-gpt models like I I didn't play last time was three months ago and I just it just it was too much work to get the right output that that I felt like oh you know also I'm building kind of on the side while I run a venture fund so my building time is limited so I try to maximize my build time I try to tackle areas that I know I can have impact quickly so I think that's an answer yeah yeah what I mean what I'm hearing from most people is that oh you can get pretty close with the other models but it's not quite there yet it's good if it's for a very specific function you can kind of like fine tune it for that but that's that's what I'm hearing from

people so just having tackled it on it yo hey you you mentioned some of the tools that you kind of used and created but now in modern times like after a year of this being in existence or two years what should people look into what tools should they use or what areas should people look into to make a difference or to to do anything interesting with AI I mean I think these large language models are are definitely something you want to learn how to use what I tell people is you know if you haven't yet start by using chat GPT because it's the lowest hanging fruit it's easy the chat interface is a nice introduction

um if you want to learn to build uh after chat GPT I tell people to go to the playground that's I think platform.openai.com which is where you can kind of play in a web UI directly with the raw API and it kind of gives you a sense of what it feels like to play with it because you can kind of update the parameters you can use a web UI to engage with basically the API directly this will give you a sense of kind of what it's like you know what you're sending into the API and once you have that and even today I play with the playground to kind of test my API calls then you can start pulling that into an API call and then wrap it with your code and start kind of building a framework around it whether you want to serve it

as a web UI create an API with it or whatnot and as you want to learn about how to kind of connect these some tools I suggest to learn how to use would be I mean Lang Sheen I think is just a really good introductory tool to start learning about kind of how agents work I would say llama index is another one if you're one interested in kind of memory kind of you know or or you dealing with lots of data you know retrieval from the larger data set llama index is is kind of the kind of a big one in that space and then if you want to learn how to build autonomous agents obviously I'm biased but I highly suggest digging into the baby AGI code that's cool by the way is there anything else planned for baby ji that you're

working on I think I've seen baby b AGI right so bbagi you know was 105 lines of code yeah so of course we got a lot of pull requests I and I don't know how to use GitHub so um uh frequency from the community was kind enough to jump in and offer to look at the pull requests and support pulling them in so we've started adding a lot of functionality of the core baby AGI unfortunately I'm actually not really good at reading other people's code and it just doesn't work with my workflow which is I need to copy paste the whole script into chat GPT so I start I I started modding the original commit that I did and so baby AGI the OG baby Asia 100 lines of code has been modded into baby b AGI baby cat AGI and then baby

deer AGI and then each time I add a couple of functions and really I'm experimenting with the core framework so it's less about adding new tools but like let's just you know it's easier to rip everything apart when there's no integration so that's that's kind of what I've been doing which is trying to figure out what the right framework for task management is um and then so um right now last night I was working on my next mod of that it's it's an alphabetical order right baby AGI be cat deer so the next one will start with E nice well yo hey thank you so much for coming on the show and sharing with us kind of your story and the story of baby AGI like having this chat about this kind of booming space thank you so much thank

you so much for having me I want to say thank you to the podcast team Henrik Jonas Rachel and Tanya and to our listeners thanks for listening