# The Future of AI Project Funding - Anna Kaic | General Magic/Giveth

- Channel: [ETH Belgrade Community](https://streameth.org/eth-belgrade-community)
- Date: 2025-10-07
- Duration: 17:20
- Topics: People & Blogs
- Watch: https://streameth.org/watch/yt-6C5GrTfokBM
- YouTube: https://www.youtube.com/watch?v=6C5GrTfokBM

## Description

The Future of AI Project Funding - Anna Kaic | General Magic/Giveth

## Transcript

The last speaker was very calm. I'm actually very nervous every time I speak in the Balkans because I feel like I'm home, so I have to perform better. Um, but I would like to share with you a little bit more about the future of AI funding. Um I've been working as a grants manager for the past year at uh General Magic and it has been a wonderful experience. But in the past six months we've seen a surge. Everybody wants to give a grant for those who build something with AI. So in case you're building something with AI or you're planning to um we are going to see how to skim through all the opportunities that are out there so you can go um yeah be safe of analysis paralysis. Um and after that we're going also to go through some practical examples of good grants, good opportunities that are out there. And uh yeah, so I've I'm from Pula, Croatia, by the way. I've been working also as a marketing manager. So we'll be talking a little bit also about communication and signaling your project out there so it actually gets used by uh the final user. But first of all, something may be obvious, some things may be obvious, but I wanted to go through everything as much information as possible. Uh something that we have seen um with other grants like traditional grants and opportunities is that they often overlook decentralized AI initiatives like in case you are not building something I don't know for for profit or you are not a closed company then funding is not for you. Uh but this is why web3 grants are great for for your pro project and we have seen many blockchains whoop we have seen many blockchains uh launching actually grant streams that try to bootstrap AI innovation on chain and the most valuable thing that I have seen happen with these uh grants is that they don't only offer money so take the money and go do whatever you want they also offer offer mentorship, they offer co-arketing uh opportunities, they offer community exposure that a lot of times are overlooked specifically by teams that are like focusing heavily on the developer side. So when you have your project, you are out there, you're looking for funds and you write AI web tree grants, you are going to find thousands of them. There are so many of them out there. But from my experience as a grants manager, how we filter grants for our clients at General Magic. Um this is more or less a cheat sheet uh that I'm sharing with you. Um it's public information but it's in one space now uh in one place now. So the first thing we do and the first thing you will notice when you look for grants for AI projects, the pools are immense. So we are talking in millions. We're talking sometimes in billions. we will later see a program that has like I think 10 billion dollars uh to to be allocated. Um but those numbers are not what you should be looking for because of course they're not going going to allocate it just to you. Um what you should be looking for is the per project cap and you have to have a clear idea how much is it going to cost for you to create your project to pay the people to actually launch it to do all the marketing things um and to actually have a successful um AI project. So once you have an idea of your cost then you can you know filter out the grants that are smaller. We will see later some bigger grants, smaller grants but this should be like the first step. If you don't have enough money don't go for the uh if you cannot ask enough money don't go for that grant. After that um what are you focusing on? Is your project purely focused on AI or are you doing a mix of you know you have a general web three uh project and then you want to implement AI in it. uh you need to be very clear on the eligibility criteria and this like take your time on this because it has happened to me as well that I am halfway through writing a grant and then there is one line that is an eligibility factor that yeah you need to just throw away everything and look for another grant. Um, something that is also important specifically for your overwhelmed and not feeling frustrated while uh waiting for funding is determine if you want to go for a time limited round or a rolling base round. Rolling base rounds are somehow a bit loose. Sometimes they are out there. They don't have a deadline. You can apply. You have a form, but then you don't know when are you going to hear back? Are you going to hear back? Are they still active? So if you don't have a direct contact with the grant allocator and you can ask hey is this still on like am I going to hear back from you in the next couple of weeks I wouldn't really go for uh rolling basis grants but time limited one are great they have a deadline they normally tell you hey it's going to take us three weeks to see who submitted and then we're going to have a decision by this date uh and in this way you can plan you know you can um plan when are you going to get the funds what are you going to do and also I would check If it's milestone based grant, so when you deliver a milestone, you get paid. If it's a retroactive grant, if they give you the whole amount immediately, which doesn't really happen uh that much because it's risky. And last but not least, as I said before, the most valuable thing specifically for new projects and specifically for new builders, like we're going to have a hackathon here, uh fresh new minds coming into the space. It's great to be able to tap into mentorship and technical support and the community behind I don't know if you are doing um an AI agent on polygon it would be great to get mentorship on polygon specifically if you're just starting out in the space you don't know anyone you don't know what to do so these are richer grants like they don't offer only funding they offer the whole package okay now uh we're going to talk about money and about some programs that um are out there. Some of them are not active in this very moment but we have seen them going on and off periodically. Uh so it's good to know which platforms or yeah which websites to be checking um and yeah to be aware of when these grants opportunities become uh available. So the first one it's the arbitrum foundation trailblazer AI grant program. Um they have 1 million total pool and this is the same thing we were talking about. So they want to allocate 1 million um but they are going actually to allocate up to 10k for projects that are already launched on arbitum. This is the key eligibility criteria. Uh and uh they have to have a live AI agent integrated. So this is good because now you know what they are going to be looking for. So if you have some spare time, you can already start working on something like this and then when the grant program opens again, you're eligible, you get funding and great. um you need to demonstrate uh an innovative usage of the uh AI agent and also adoption potential because pumping out AI agent without usage is not really useful for anyone. And yeah, the future rounds are going to be announced on the website. So stay tuned with Arbitron Foundation. If you're a fan of Avalanche Foundation, they have the infraidal AI program. uh again they're allocated 50 million indirect funding for uh of retroactive grants. So in case you have already built something uh on avalanche it's uh good to know so that when this they have periodic pitch days where you can go and say hey I built this and then you get funding the next pitch day is tomorrow so it's a bit rushed but uh they do it they do it very frequently so if you stay um you know up to date with their community and it's always very useful to um specifically for these grants to join their discord join their telegram be able to ask questions because once you get into applying for a grant, you will have many questions because some questions in the grant are not that clear. So in case you can get a direct contact, it's very good to get involved with the community. So I would say pick one program that is interesting for you. Really get interested in it. Your chances for getting a grant will be uh higher. And this one is very interesting because they actually implemented an AI agent inside the program called Alfred. Interesting name. Um and it helps evaluating the project. Uh it's not the only way they use to evaluate projects. Uh but it's a cool mixture between AI and human evaluation. Um this is one that is very fun because they had a memecoin contest as well going on like couple of weeks ago and they were allocating 1 million poll um with through the polygon community grants. So if you go on their dashboard, you will see they have many tracks. Uh for us right now, the AI track is uh really interesting. They're going to allocate 1 billion pole in the next 10 uh 10 years. So which means 10 million uh pole every year, which is amazing. Um despite the price of Paul, we're not talking about that now. Uh, but it's great for um, yeah, for a for builders who want to stay loyal to Polygon, build on Polygon. And here I put some examples of projects that already got great funding there. So you can look up what they have done so far. Um, your project must be uh, built within Polygon of course and uh, so EVM compatible and as something we have seen in the past programs as well, it needs to be innovative like don't copy paste from other projects. Uh some a program that is not really that popular at least I didn't hear that much about it is the light chain protocol. Um and I think it's because they are offering a relatively smaller pool. Um but like 5K per team is not that impressive or that much but I think with AI you can build uh you can build something cool with uh 5K. This is a great program if you're starting uh because they offer technical men mentorship for the uh core team. Uh so this might be you know a good opportunity maybe for hackers that will be here at Belvelgrid. Um and 5K should be relatively easy to get as long as you comply with all the requirements and they're going to help you also with co-arketing and community events and you're going to get out there and your product will be uh actually uh actually used. So if you are starting with uh yeah AI based project uh check out light chain. Then for all the lovers of like regenerative finance and real world assets and everything in between, uh, Kadada has a great program, 50 million uh, dollars in uh, funding. Um, and basically it's it's an allocation that I think it's more prone to projects who want to have a real world usage in the meaning that you know maybe tracking um, tree planting. I've been in regenerative finance for some time and it's very difficult to find projects that are supportive of such initiatives. Uh so Kadena has been doing a great job with uh with these type of funings and uh yeah you can apply anytime. So this is the rolling basis uh thing I was telling you. Um before you can apply anytime I think you can you will probably hear back from them in a couple of weeks. Then uh for internet computer fans uh they have the V AI grant. Uh this is also a rolling basis grant but it's good because they give you a timeline where you can hear back from them but as you can see it's six weeks. So if you cannot wait for six weeks then don't go here. Uh but again this is if you want to build on um internet computer and uh basically yeah it's it might be easier to get a grant like this just because the amount of people will be yeah the amount of project and the amount of competition will be uh will be less. And last but not least uh we have the Solana Foundation AI grant and also partnership. This is also on a rolling basis, but this is an established program that has been going on for a long time. So if you decide to go with Solana, you will hear back from them for sure. Um, and the focus of course is integrating AI with the Solana blockchain and you need to be open source. And uh here I put some collaborations that they have done like they had a collaboration with grant uh with sorry with art arc and um yeah this one this one was rust based AI agent development uh but they have been having different partnerships uh on and off. So in case yeah you are specifically interested in a specific collaboration just stay tuned. They have had so many that yeah you will find something for you as well. Okay, now you we have talked about different programs. We have seen programs that are for beginners, for pro for teams that are maybe a bit bigger, rolling bases, closed uh grants with a deadline. We have seen all of that. Um but I think the when you will actually uh sit down and write your grant, these are the things I would really focus on. Um specifically because you want to be straightforward as much as possible, answer to all the questions, but avoid being too concise. Um so first of all, it's quite obvious like you need to be uh complying with the onchain criteria. If they want you to build on Polygon, you need to build on Polygon and and vice versa for other um other blockchains. The clear milestones and deliverables are going to be your best friend. Like if you don't have So one one important thing, you don't need to determine the milestone for the whole project. If you can get funding for building just one part of your um AI project, it's fine as long as you can ship it and as long as you can prove like it it can be enough to prove that you have done something. Um but defining milestones for one part of it is important and defining the milestones for the whole product is important. So that sometimes it happens that projects cannot get straight like 150k of funding immediately. So they apply with three different grants, but having milestones, it's the most important part. They're going to probably um give you the funding after you complete a milestone. So just be clear on that. And very very important, you have to have some community. You have to be known. You have to talk to people. You they like grant allocators don't really like anons. They don't like people who just come to their community and they're like, "Yeah, I decided I'm a really like a Solana fan, so give me money." try to be involved, try to be present and try to be really like aligned with the blockchain you decide to bre to to build on. But uh yeah, that is uh quite obvious but it doesn't hurt to mention and yeah one important thing make sure if they want open source that you are actually open source and you don't write the whole grant and then you waste your time. Um so next steps for you if you want to make sure that you don't miss any of these opportunities um you can bookmark each of these uh different yeah uh blockchains that we mentioned today. The if you are early it's better because applying like right before the deadline will not get give you the will not give the chance to grant reviewers to come back to you and ask for more proof more documentation discuss better what you want to build. And this is something that really happens like it's not just I will write my form, I will um submit it and I will get a grant. There is a back and forth between you and the grant allocator that you want to make sure is done properly specifically if you ask like for a lot of money. Um yeah, I that was my part. It was uh quite quick but uh and I see I have a QR code here that the the one on the right on your right says general magic. I it's the project I work uh at at the moment. We offer grants as a service uh for our clients as well and uh because there are teams who don't really have time to do the whole research and then matchmake and then write the grant. So if you're interested in this or other services because we have many services you can follow us on X. Okay, done. I thank you.