What If AI Models Lived Onchain? Exploring a Bold New Frontier with Arbitrum Stylus

Proposals
1

Abstract

This proposal is developing a framework for deploying and interacting with AI models onchain using Arbitrum Stylus and Rust. This initiative will unlock innovative use cases for the Arbitrum ecosystem, such as decentralized AI computation and verifiable machine learning, while showcasing the scalability and efficiency of Stylus. By focusing heavily on development resources and open-source contributions, this project aims to establish Arbitrum as a hub for decentralized AI applications.

Motivation

The idea for this proposal emerged from a grassroots conversation within the Arbitrum communities. A Twitter thread sparked by Patrick McCorry’s observations on the EthGlobal hackathon led to a comment by Matt Hamilton (Senior DevRel at Arbitrum) suggesting embedding AI neural networks in NFTs using Stylus. Inspired by this, I conducted an experiment deploying a decision tree model using Stylus and Rust, which proved functional but highlighted challenges.

Having previously and currently contributing to the Stylus ecosystem through the @stylusdev Twitter account, where I create developer education content, I am passionate about pushing Stylus’s capabilities and onboarding developers. This proposal is the next logical step in realizing Stylus’s potential by addressing existing limitations and building tools to make onchain AI a reality.

The motivation is clear: onchain AI models represent a groundbreaking use case for blockchain technology, enabling trustless, verifiable, and decentralized computation for industries such as finance, gaming, and governance.

Rationale

This proposal aligns closely with the Arbitrum DAO’s mission and guiding values:

  1. Innovation: Stylus is a cutting-edge platform, and this project showcases its flexibility and scalability for high-performance computation.
  2. Ecosystem Growth: By introducing onchain AI capabilities, Arbitrum can attract new developers and projects to its ecosystem, expanding its use cases beyond DeFi and gaming.
  3. Community Engagement: The project will release open-source tools and developer education materials to empower the broader Arbitrum community.

As a Rust developer with deep experience in Stylus, I have already laid the groundwork for this initiative. The project will also leverage the expertise of two ML engineers and another Rust engineer to ensure scalability and successful delivery.

Specifications

Platform and Technologies

  • Rust and Stylus: Chosen for their performance and compatibility with WebAssembly, enabling computational efficiency for onchain AI.
  • Arbitrum Stylus Layer 2: Leveraging its scalability and low gas costs for deploying and interacting with AI models.
  • zk-Proofs: Potential integration for verifiable and privacy-preserving AI inference.

Design Decisions

  • Lightweight AI Models: Initial focus on simple models to ensure feasibility as this is a research initiative.
  • Modular Architecture: Developing reusable libraries for Rust and Stylus, allowing easy adoption by other developers.

Steps to Implement

  1. Phase 1: Research and Feasibility (Months 1)
  • Conduct a detailed study on the scalability of Stylus for AI models.
  • Identify the most suitable lightweight AI models for deployment.
  • Begin integrating zk-proof mechanisms.
  1. Phase 2: Prototype Development (Months 2-3)
  • Build and deploy a prototype AI model (e.g., decision tree) on Arbitrum Stylus.
  • Test performance metrics such as gas usage and inference speed.
  • Publish a feasibility report.
  1. Phase 3: Optimization and Expansion (Months 4)
  • Optimize the framework for larger datasets and more complex models.
  • Incorporate community feedback and refine the tools.
  • Build open-source Rust libraries for broader adoption.
  1. Phase 4: Community Engagement and Finalization (Months 5-6)
  • Host workshops and publish developer guides/tutorials.
  • Open-source the complete framework.
  • Submit a final report on milestones achieved and future directions.

Timeline

  • 26th November 2024: Proposal finalized and RFC submitted.
  • 2nd December 2024: Snapshot proposal submitted.
  • 2nd - 9th December 2024: Snapshot voting period.
  • 9th - 16th December 2024: Tally voting and final confirmation.
  • 20th - 23rd December 2024: Project kick-off and team onboarding.
  • ** January - June 2025:** Rust and Stylus Mode (Arbitrum Hat On)

Overall Cost

Allocation and Justification

1. Cloud Computing and Storage ($80,000)

Justification: Training AI models, even lightweight ones, can be computationally intensive and require significant resources to achieve reasonable training times and performance.

2. Data Acquisition and Preparation ($30,000)

Justification: High-quality data is crucial for training effective AI models. Costs cover acquisition and preparation to ensure models are trained on reliable data.

3. Software Licenses and Tools ($10,000)

Justification: Specialized tools can significantly enhance development efficiency, model performance, and security, justifying their cost.

4. Security Audits and Code Reviews ($20,000)

Justification: Security is paramount, especially when deploying onchain. Audits help prevent costly exploits and build trust with users and the community.

5. Research and Development Expenses ($30,000)

Justification: Ongoing research is essential to push the boundaries of what’s possible and to incorporate the latest advancements into the project.

6. Testing and Deployment ($10,000)

Justification: Robust testing and deployment processes ensure reliability and facilitate smooth integration and updates.

7. Community Tools and Documentation ($10,000)

Justification: Comprehensive documentation and support tools are vital for community adoption and to lower the barrier to entry for new developers.

8. Intellectual Property and Legal ($5,000)

Justification: Proper legal groundwork prevents future disputes and ensures the project adheres to all regulations and best practices.

9. Contingency Funds ($5,000)

Justification: A contingency fund ensures the project can handle unexpected challenges without compromising its goals.

TOTAL: 200, 000 USD

What Success Looks Like

  • Deployment of a working onchain AI model prototype on Stylus.
  • Release of open-source tools and libraries for the community.
  • Increased developer interest in Stylus and Rust within the Arbitrum ecosystem.
  • A roadmap for future expansion into more complex models and decentralized AI applications.

Conclusion

This proposal is an opportunity to push the boundaries of what blockchain technology can achieve by enabling decentralized AI computation on Arbitrum Stylus. By combining a solid technical foundation, community engagement, and a passion for innovation, we aim to make Arbitrum the go-to ecosystem for onchain AI applications.

1 Like
2

Hey there @0xredacted !

Did you followed what happened in this Stylus hackathon, 2 weeks before devcon, by any chance?

The winning project of the hackathon actually tried to do this. They called themselves The Real Open AI. =)

1 Like
3

I have a few questions due to the fact that I am not very familiar with AI and how to use it in crypto.
1.

What does this mean? Scalability in what sense, I did not quite understand what the result should be.
2.

What are the options, how can we use AI models in our case?
3.

Arbitrum does not have this now, why do we need to do this here?
4.

That is, our task is to create a framework for creating AI?
I just did not find this in the text earlier.

Standard questions about the budget:

  1. How many people and hours are required for each type of work.
  2. What is the hourly wage
  3. Security audit for which product?
    If it is a framework, then the audit must be carried out for the result in any case.
4

Have you checked out the Trailbrazer AI Grant Program? The Arbitrum Foundation just announced it.

Introducing the Trailblazer AI Grant Program—a $1,000,000 fund aimed at empowering visionary developers to create cutting-edge AI agents and onchain AI products on Arbitrum chains.

image
image1600Ă—895 83.2 KB

1 Like
5

Hello @paulofonseca.

Yes, I did follow the Stylus hackathon and was impressed by “The Real Open AI” project. Their work showcased the exciting possibilities of bringing AI models onchain using Arbitrum Stylus. Their approach to democratizing AI and involving the community through NFTs is inspiring and highlights the innovative spirit within the Arbitrum ecosystem.

My proposal aligns with this vision but takes a different approach by focusing on building the foundational infrastructure for onchain AI development. Instead of developing a single application or model, we aim to create a comprehensive framework, including tools, libraries, and documentation, that will make it easier for anyone to build and deploy AI models on Stylus using Rust.

By concentrating on the underlying infrastructure, we hope to empower developers to explore a wide range of AI applications onchain, from simple models to more complex systems. This will not only broaden the use cases for Arbitrum Stylus but also foster a community of innovators who can contribute to the growth and diversification of the ecosystem.

In essence, while “The Real Open AI” project demonstrated the potential of onchain AI through a specific application, our proposal seeks to expand that potential by providing the resources and infrastructure necessary for widespread adoption and development of onchain AI models on Stylus.

1 Like
6

Hello @cp0x

Great question! When we refer to the “scalability of Stylus for AI models,” we’re talking about the ability of the Arbitrum Stylus platform to efficiently handle the deployment and execution of AI models on the blockchain, especially as the complexity of the models or the number of users increases.

Here’s what that means in more detail:

1. Computational Efficiency

  • Understanding the Challenge:
    • AI models, even simple ones, require computational resources to run (perform inference) or to train.
    • Blockchains are inherently resource-constrained environments compared to traditional computing platforms.
  • Scalability Aspect:
    • Scalable in Terms of Performance: Ensuring that AI models can run efficiently without causing network slowdowns or excessive gas costs.
    • Optimizing Resource Usage: Finding ways to optimize AI computations so they fit within the blockchain’s limits.

2. Handling Larger Models and Data

  • Understanding the Challenge:
    • More complex AI models (like deep neural networks) involve large amounts of data and numerous calculations.
    • Running these directly onchain can be impractical due to limitations in processing power and storage.
  • Scalability Aspect:
    • Model Optimization: Adapting AI models to be lighter and more efficient so they can run onchain.
    • Layered Approach: Possibly using offchain resources for heavy computations and onchain for verification and smaller tasks.
    • Supporting Growth: Ensuring that as models become more sophisticated, the platform can still support them effectively.

3. User and Developer Growth

  • Understanding the Challenge:
    • As more developers start deploying AI models and more users interact with them, the network could experience increased load.
  • Scalability Aspect:
    • Network Throughput: Stylus needs to handle more transactions and computational tasks without congestion.
    • Cost Efficiency: Keeping gas fees reasonable so that increased usage doesn’t make the platform prohibitively expensive.

4. Practicality and Cost-Effectiveness

  • Understanding the Challenge:
    • Running computations on the blockchain can be expensive due to gas fees associated with processing power.
  • Scalability Aspect:
    • Reducing Gas Costs: Optimizing smart contracts and computations to minimize the gas required.
    • Efficient Code Execution: Leveraging Rust and Stylus to write more efficient code that executes faster and cheaper.

5. Security and Reliability

  • Understanding the Challenge:
    • As models and usage scale, ensuring the system remains secure and reliable is crucial.
  • Scalability Aspect:
    • Robust Architecture: Building the framework to handle more operations without compromising security.
    • Error Handling: Ensuring that increased complexity doesn’t lead to more vulnerabilities or failures.

Intended Result

  • Demonstrate Feasibility: Show that it’s practical to deploy and run AI models on Arbitrum Stylus without prohibitive costs or performance issues.
  • Enable Complex Applications: Allow developers to create more sophisticated decentralized applications (dApps) that leverage AI capabilities on the blockchain.
  • Foster Ecosystem Growth: By proving scalability, we encourage more developers to build on Stylus, leading to a richer and more diverse ecosystem.

In Simple Terms

  • Scalability in this context means making sure that the platform can handle:
    • More Complex AI Models: Not just simple algorithms, but increasingly sophisticated ones.
    • More Users and Developers: As the number of participants grows, the system continues to perform well.
    • Increased Computational Load: The platform can process more data and calculations without slowing down or becoming too expensive.
2 Likes
7

again @cp0x

Great question! When we mention “identifying the most suitable lightweight AI models for deployment,” we’re focusing on selecting AI models that are efficient, require minimal computational resources, and can operate effectively within the constraints of a blockchain environment like Arbitrum Stylus. Here’s a detailed explanation of the options and how we can use AI models in our case.

Options for Lightweight AI Models

1. Linear Models

  • Linear Regression
    • Use Case: Predicting continuous numerical values.
    • Why Suitable: Involves simple mathematical operations; low computational overhead.
    • Example Application: Predicting future asset prices based on historical data.
  • Logistic Regression
    • Use Case: Binary classification problems (e.g., yes/no decisions).
    • Why Suitable: Computationally efficient; outputs probabilities.
    • Example Application: Determining the likelihood of a transaction being fraudulent.

2. Decision Trees

  • Use Case: Classification and regression tasks.
  • Why Suitable: Based on simple conditional (if-else) statements; depth can be limited to control complexity.
  • Example Application: Automated decision-making processes, such as credit approvals.

3. Naive Bayes Classifier

  • Use Case: Classification tasks, especially with text data.
  • Why Suitable: Relies on calculating probabilities; computationally light.
  • Example Application: Categorizing messages or transactions as spam or legitimate.

4. Rule-Based Systems

  • Use Case: Systems based on predefined rules.
  • Why Suitable: Highly transparent; easy to implement with basic logic.
  • Example Application: Smart contracts enforcing regulatory compliance.

5. Simple Neural Networks

  • Use Case: Basic pattern recognition and function approximation.
  • Why Suitable: Small networks (e.g., one hidden layer) can be manageable onchain.
  • Example Application: Simple predictive models for user behavior.

How We Can Use AI Models in Our Case

Our goal is to develop a framework that allows developers to deploy and interact with AI models on Arbitrum Stylus. Here’s how we can use these AI models:

1. Onchain Inference

  • What It Means:
    • Running the AI model directly within a smart contract to make predictions or decisions.
  • Application:
    • Smart contracts can automatically execute actions based on model outputs.
  • Example:
    • An insurance contract that assesses claims automatically using an onchain model.

2. Automated Decision-Making in Smart Contracts

  • What It Means:
    • Embedding AI models to handle conditional logic more intelligently than static rules.
  • Application:
    • Dynamic adjustment of interest rates in lending platforms based on risk assessment.
  • Example:
    • A lending protocol that adjusts collateral requirements based on borrower profiles.

3. Enhancing NFTs with AI

  • What It Means:
    • Integrating AI models into NFTs to enable dynamic behavior.
  • Application:
    • NFTs that evolve over time or in response to certain conditions.
  • Example:
    • A virtual pet NFT that “learns” and changes based on owner interactions.

4. Verifiable Computations

  • What It Means:
    • Performing computations where the correctness can be verified by anyone.
  • Application:
    • Using zero-knowledge proofs to verify offchain model training results onchain.
  • Example:
    • A prediction market that verifies the integrity of submitted predictions.

Use Cases Specific to Our Proposal

1. Decentralized Finance (DeFi)

  • Risk Assessment:
    • Use logistic regression to evaluate the risk level of loan applicants.
  • Price Prediction:
    • Linear regression models to forecast asset prices for automated trading strategies.

2. Supply Chain Management

  • Verification Systems:
    • Decision trees to verify product authenticity and traceability.
  • Automated Compliance:
    • Rule-based systems to ensure all transactions meet regulatory standards.

3. Gaming and NFTs

  • Dynamic NFTs:
    • Simple neural networks to enable NFTs that evolve based on gameplay or user interactions.
  • Game Mechanics:
    • AI models to control non-player characters (NPCs) behavior onchain.

4. Identity and Reputation Systems

  • Trust Scores:
    • Naive Bayes classifiers to compute user reputation scores based on onchain activity.
  • Access Control:
    • Decision trees to grant or restrict access to certain features or services.
1 Like
8

@cp0x

Even though Arbitrum doesn’t currently have built-in support for zero-knowledge proofs (zk-proofs), integrating them into our project is important for several reasons:

  1. Enhancing Privacy and Security:
  • Protect Sensitive Data: zk-proofs allow us to verify that computations (like AI model results) are correct without revealing the actual data or details. This means we can keep sensitive information private while still ensuring trust.
  1. Trustless Offchain Computation:
  • Efficient Processing: AI models can be complex and resource-intensive. By performing heavy computations off the blockchain and using zk-proofs, we can prove to the blockchain that the computations were done correctly without overloading it.
  1. Future-Proofing the Project:
  • Staying Ahead: The blockchain industry is moving toward more privacy and efficiency, with zk-proofs becoming more common. By integrating them now, we’re preparing our project to be compatible with future updates and advancements in Arbitrum and other platforms.
  1. Improving Scalability:
  • Handling Growth: zk-proofs help us keep the on-chain transactions light and efficient, which is essential as more users and developers start using our framework.

In Simple Terms:

  • Why Include zk-Proofs Now? Even if Arbitrum doesn’t support them natively yet, using zk-proofs makes our project more secure, private, and ready for the future. It allows us to handle complex AI tasks more efficiently by verifying work done off-chain without burdening the blockchain.
1 Like
9

Stylus’s ability to bring Arbitrum closer to a wider pool of engineers and developers is a key competitive advantage for the ecosystem. We believe this inclusivity and versatility will help grow the ecosystem and attract more innovation and we are excited to see initiatives that go in that direction.
This is an area where we should continue to invest and promote, as it positions Arbitrum uniquely in the ecosystem by attracting diverse talent and expanding its use cases.

Specifically on this proposal, the use cases you’ve highlighted, particularly around decentralized AI decision-making and smart contract automation, are very interesting. These applications demonstrate the transformative potential of integrating AI directly with smart contracts and showcase Arbitrum’s ability to lead in this innovative space.

Because of these, we are generally supportive of the proposal. It balances bold ambition with a practical plan. We appreciate and want to highlight the importance of redirecting funds to other initiatives if the project is deemed unfeasible in Phase 2. It ensures we encourage big ideas, while making sure DAO resources are used effectively.

10

Yes, you’re correct! One of the primary goals of our proposal is to create a framework that enables developers to deploy and interact with AI models onchain using Arbitrum Stylus and Rust. Our task goes beyond implementing a single AI model; we’re building the foundational infrastructure—including tools, libraries, and documentation—that will make it easier for anyone to create and optimize AI models for onchain deployment.

1 Like
11

Answers to Your Specific Questions

1. How many people and hours are required for each type of work?

  • Data Acquisition and Preparation:
    • Personnel: 2 Data Engineers/Data Scientists
    • Hours per Person: 200 hours
    • Total Hours: 400 hours
  • Research and Development:
    • Personnel: 2 Machine Learning Engineers
    • Hours per Person: 250 hours
    • Total Hours: 500 hours
  • Testing and Deployment:
    • Personnel: 2 QA/Test Engineers
    • Hours per Person: 100 hours
    • Total Hours: 200 hours
  • Community Tools and Documentation:
    • Personnel: 1 Technical Writer/Developer Advocate
    • Hours: 200 hours
      4. Security Audits and Code Reviews ($20,000)

Product to be Audited: The on-chain AI framework/infrastructure, including smart contracts, Rust libraries, and any deployed code.

Personnel:

  • External Security Auditors: 2 auditors (from a reputable firm).
  • Estimated Hours per Auditor: 170 hours.
  • Hourly Rate: $58.82/hour.

Total Cost Calculation:

  • Total Hours: 2 auditors * 170 hours/auditor = 340 hours.
  • Total Cost: 2 auditors * 170 hours/auditor * $58.82/hour = $20,000.
  • Total Project Hours: 1,460 hours

2. What is the hourly wage?

  • Data Engineers/Data Scientists: $50/hour
  • Machine Learning Engineers: $60/hour
  • QA/Test Engineers: $50/hour
  • Technical Writer/Developer Advocate: $50/hour
  • External Security Auditors: $59/hour

3. Security audit for which product?

  • Product: The onchain AI framework, which includes:
    • Smart Contracts: Developed using Rust and Stylus, handling AI model deployment and interaction.
    • Rust Libraries and Tools: Supporting libraries for AI integration with Stylus.
    • Any Deployed Code: Associated with the framework.
  • Purpose of Audit: To ensure the security and reliability of the framework before public release.

4. If it is a framework, then the audit must be carried out for the result in any case.

  • Confirmation: Yes, the audit is essential and will be conducted on the final framework to identify and fix any vulnerabilities. This is accounted for in the budget under “Security Audits and Code Reviews” with an allocation of $20,000.
1 Like
12

Wow @karpatkey, thanks so much for your amazing feedback! I’m super excited that you’re as hyped about Stylus as I am. Bringing more engineers and developers into the Arbitrum ecosystem is such a game-changer, and it’s awesome to see others sharing that vision.

We’re just scratching the surface. Can’t wait to see where this innovation takes us and how Arbitrum can lead the charge.

Really appreciate your support on the proposal. Balancing ambitious dreams with practical steps is key, We’re all about making bold moves while ensuring the DAO’s resources are used wisely.

Thanks again for the encouragement! Let’s push the boundaries together and make some magic happen!

1 Like
13

This seems like a potentially worthwhile business, at least that’s how you’re selling it to the DAO. I think you should look at venture funding for something like this rather than asking the DAO. I’ll be voting no on this proposal.

14

Hello @NathanVDH,

Thank you for sharing your thoughts. I appreciate your perspective and the opportunity to clarify our intentions.

Our primary goal with this proposal is to contribute to the Arbitrum ecosystem by building open-source infrastructure that benefits the entire community. By developing a framework for deploying AI models onchain using Stylus and Rust, we’re aiming to create tools and resources that any developer can use to innovate and expand the capabilities of decentralized applications on Arbitrum.

This project isn’t about establishing a proprietary business or generating profits—it’s about fostering innovation within the community. The funds requested are allocated toward development costs, infrastructure, and creating open-source tools that will be freely available to everyone in the ecosystem.

Seeking venture funding often comes with expectations of proprietary ownership and a focus on financial returns, which might limit accessibility and openness. By collaborating with the DAO, we ensure that the project’s outcomes remain community-driven and aligned with the collective interests of Arbitrum users and developers.

I understand your concerns, and I hope this explanation provides clarity on why we’re seeking support from the DAO rather than venture capital. I’m open to further discussion if you have additional questions or suggestions. Your feedback is valuable, and I appreciate you taking the time to engage with the proposal.

1 Like
15

As Stylus is a key competitive advantage for Arbitrum, I think the DAO should encourage initiatives leveraging it. Because of this, I am happy to see this proposal.

Generally, I am supportive of it. But, I’d like to ask a clarification on funding. Do you expect to seek funding from other places as well or would this be funded by Arbitrum DAO only?
For example, the Arbitrum Foundation has recently launched the Trailblazer grant program, specifically focused on powering AI innovation on Arbitrum. This project might be a good fit for it.

1 Like
16

Hello @0xDonPepe

Yes, I’ve seen the Trailblazer AI Grant Program, and it’s a fantastic initiative for supporting AI projects within the community. However, the program has a cap of $10,000 per project, which isn’t sufficient for the scale of our proposal. Our project involves significant research and development to build onchain infrastructure for AI on Arbitrum. This includes developing a comprehensive framework, tools, and open-source resources that require a larger budget to execute effectively. That’s why we’re seeking funding through this proposal to ensure we can deliver meaningful results that benefit the entire ecosystem.

17

Hi @alicecorsini

Thank you for your support and for recognizing Stylus as a key competitive advantage for Arbitrum! We’re excited to contribute to leveraging its potential within the ecosystem.

Regarding your question about funding, we’re currently seeking support solely from the Arbitrum DAO for this project. The scope of our proposal involves significant research and development to build comprehensive onchain AI infrastructure using Stylus and Rust. This includes developing open-source tools, frameworks, and documentation that will benefit the entire Arbitrum community.

We have considered the Trailblazer AI Grant Program, and we think it’s a fantastic initiative. However, the program has a funding cap of $10,000 per project, which isn’t sufficient for the scale and ambitions of our proposal. Our project requires a larger budget to cover the extensive development efforts and resources needed to achieve our goals.

By seeking funding from the DAO, we aim to ensure that the project’s outcomes align closely with the community’s interests and that we can deliver meaningful contributions to the ecosystem. We believe this approach allows us to make a more significant impact and fully realize the potential of integrating AI with Stylus on Arbitrum.

Thank you again for your support!

18

Hello! Thanks for your proposal!

While it seems a nice idea, I have a few questions about the execution:

  • At the end of phase 2, you mention the publishing of a feasibility report. What would happen if the results so far show that it is not feasible to continue?

  • Who will decide if it is feasible or not? Is there any external entity taking part in this?

  • Could you break down the budget/cost to match the implementation steps?

  • Would you be open to receive the funds as a stream, instead of the full amount upfront?

19

Hello @jameskbh.

Thanks for the compliment and questions, below find the reponses

1. At the end of Phase 2, you mention the publishing of a feasibility report. What would happen if the results so far show that it is not feasible to continue?

If the feasibility report at the end of Phase 2 concludes that the project is not viable to continue, the unspent funds would be returned to the DAO or redirected to other community initiatives based on a transparent decision process. This ensures that the resources are used responsibly and remain aligned with the interests of the Arbitrum ecosystem.

2. Who will decide if it is feasible or not? Is there any external entity taking part in this?

The decision on feasibility will be based on the findings of the feasibility report, which will include clear metrics and benchmarks to evaluate the project’s progress. The results will be shared transparently with the DAO for review, and the community will play a significant role in determining the next steps.

While no external entities are currently involved, we are open to engaging independent experts or advisors to provide an unbiased evaluation of the report if the DAO recommends it.

3. Could you break down the budget/cost to match the implementation steps?

Certainly! Here’s a breakdown aligned with the implementation phases:

  • Phase 1 (Research and Feasibility): $60,000
    • Cloud Computing and Storage: $30,000
    • Data Acquisition and Preparation: $15,000
    • Research Personnel Costs: $15,000
  • Phase 2 (Prototype Development): $70,000
    • Cloud Computing and Storage: $30,000
    • Software Licenses and Tools: $10,000
    • Development Personnel Costs: $20,000
    • Testing and Deployment: $10,000
  • Phase 3 (Optimization and Expansion): $40,000
    • Cloud Computing and Storage: $20,000
    • R&D Expenses: $10,000
    • Documentation and Community Tools: $10,000
  • Phase 4 (Final Report and Community Engagement): $30,000
    • Security Audits and Code Reviews: $20,000
    • Legal/IP and Contingency: $5,000
    • Workshops/Outreach: $5,000

4. Would you be open to receive the funds as a stream, instead of the full amount upfront?

Yes, we are open to receiving the funds as a stream rather than the full amount upfront. A phased funding approach aligns well with the project’s structure, as it allows the DAO to track progress and ensure that resources are being used effectively. The funds could be released incrementally at the start of each phase, contingent on successful completion and reporting of the previous phase. This approach provides transparency and accountability, which we fully support.

3 Likes
20

Hi! the proposal is not bad at all; it is well-intentioned, but I have some questions:

Regarding the initial challenges you mentioned:

It would be helpful if you elaborated on what those challenges were exactly. Were they specific technical limitations, such as gas costs, performance, or compatibility issues with Stylus? A deeper analysis of these problems could not only strengthen the justification for your proposal but also provide clearer direction for the project from the outset.

I understand you are aware of the recent hackathon in Bucharest, but considering that the winning team, “The Real Open AI,” attempted to address a similar use case, have you already explored how their approach compares to your proposal? For instance, what tools or models did they use, and what are the key differences in your implementation with:

?
Including an analysis of their work or even highlighting lessons learned could enrich your project and help avoid redundancies…

I am also interested in your approach to zk-Proofs, where you state:

Given the timeline and estimated budget, how do you plan to handle the complexity and cost this entails? Perhaps you could clarify whether this will be an initial implementation or just a research stage within the project. This would help to structure the phases.

Regarding the community, you mention:

This is key to success. But have you considered more proactive strategies to engage the community? Virtual hackathons, rewards for projects using your framework, or collaborations with prominent developers could be excellent ways to increase adoption and the project’s impact.

Finally, how do you plan to measure success in the implementation and adoption of these tools by the community? While you mention milestones like the

you could add more specific metrics… such as the number of active developers or projects adopting your framework within a given timeframe