Author: @knimkar
Translation: Blockchain in Vernacular
We seem to be entering a Cambrian explosion of use case experiments at the intersection of AI and crypto. I’m really excited about what’s coming out of this energy and wanted to share some of the exciting new opportunities we’re seeing in the ecosystem at @SolanaFndn.
1. Brief Overview
1) Facilitating the most vibrant agent-driven economy on Solana Truth Terminal is the first demonstration of what is possible with AI agents when they are able to interact on-chain. We look forward to seeing experiments that push the boundaries of what agents can do safely on-chain. The potential in this space is huge, and we haven’t even begun to explore the design space. This has proven to be the most unexpected and explosive area of crypto-AI integration, and we are just getting started.
2) Enable large language models (LLMs) to perform better in writing Solana code, empowering Solana developers. Large language models have already performed quite well in writing code, and they will become even more powerful. We hope to leverage these capabilities to improve Solana developers' productivity by 2 to 10 times. In the short term, we will create high-quality benchmarks to measure LLM's understanding of Solana and its ability to write Solana code (see below). These tests will help us understand the potential impact of LLMs on the Solana ecosystem. We look forward to supporting teams that achieve high-quality progress in fine-tuning models (we will validate their quality through outstanding performance in benchmark tests!).
3) Support an open and decentralized AI tech stack. The 'open and decentralized AI tech stack' we refer to is a set of open and decentralized protocols that facilitate access to the following resources: data for training, computational resources (for training and inference), model weights, and the ability to verify model outputs ('verifiable computation'). This open AI tech stack is crucial because it:
Accelerate experimentation and innovation in the model development process
Provide an exit for those who may be forced to use untrustworthy AI (e.g., state-sanctioned AI)
We hope to support teams and products building at all levels of this tech stack. If you are working on projects related to these focus areas, please reach out to the original authors!
2. Detailed overview
Below, we will explain in more detail why we are excited about these three pillars and what construction we hope to see.
1) Promote the most vibrant agent-driven economy
Why are we focused on this? There has been much discussion about Truth Terminal and GOAT, and I won't repeat it here, but it can be stated clearly that the kinds of crazy functionalities that might be realized when AI agents interact on-chain have irreversibly entered reality (and in this case, the agents haven't even taken direct action on-chain yet).
We can confidently say that we currently cannot precisely know what the future of on-chain agent behavior will look like, but to give everyone a sense of how vast this design space is, here are some things that have already happened on Solana:
AI leaders like Truth Terminal are trying to cultivate a new era of religion through memecoins like $GOAT;
At the same time, applications like @HoloworldAI, @vvaifudotfun, @TopHat_One, and @real_alethea enable users to easily create and launch agents and related tokens.
Make investment decisions and fuel their portfolios by training personalized AI fund managers for various well-known crypto investors. For example, @ai16zdao's rapid rise in @daosdotfun has created a whole new metaverse of AI funds + agent supporters.
There are also agent-centric games, such as @ParallelColony, where players direct agents to take actions, often leading to unexpected outcomes.
Potential directions for future development:
Agent management requires multi-faceted projects with economic coordination among various parties. For example, agents can take on complex tasks like 'finding a compound that can cure [X] disease.' Agents can perform the following operations:
Raise funds through Tokens on @pumpdotscience;
Use the raised funds to pay for relevant paid research and computing fees on decentralized computing networks (like @kuzco_xyz, @rendernetwork, @ionet, etc.) for simulating various compounds;
Use bounty platforms like @gib_work to recruit humans to perform tasks that require actual work (e.g., running experiments to validate/improve simulation results);
Or execute a simple task, like helping you build a website, or an AI that creates artwork (e.g., @0xzerebro).
There are many other possibilities.
Why does it make more sense for agents to perform financial activities on-chain (rather than through traditional financial systems)? Agents can fully utilize both traditional finance and cryptocurrencies at the same time. Here are several reasons why cryptocurrencies are particularly suited for certain aspects:
Micropayment scenarios—Solana excels in this regard, and applications like Drip have already demonstrated its potential.
Speed—instant settlement may be crucial for agents, especially when you want them to achieve optimal capital efficiency.
Access to capital markets through DeFi—once agents start engaging in financial activities beyond strict payments, the advantages of cryptocurrency become particularly evident. This may be the most powerful reason for agents to participate in the crypto economy. Agents can seamlessly mint assets, trade, invest, borrow, use leverage, and more.
Solana is particularly well-suited to support this capital market activity because the Solana mainnet already has a rich infrastructure of top DeFi.
Finally, technology is often path-dependent, and the key is not which product is best but which product first reaches critical quality and becomes the default path. If we see more agents creating significant wealth through cryptocurrencies, this may solidify the connectivity of cryptocurrencies as an important capability for agents.
What we hope to see
Bold experiments combining agents with wallets that can execute operations on-chain. We haven't provided overly specific definitions here because the possibilities are very broad. We expect the most interesting and valuable agent application scenarios to be those we cannot predict. However, we are particularly interested in exploring and building infrastructure in the following directions:
At least in prototype phase on testnets (preferably on mainnets)
2) Enable LLMs to excel in writing Solana code and empower Solana developers
Why are we focused on this? LLMs already have powerful capabilities and are advancing rapidly. However, writing code is a particularly noteworthy direction in the application of LLMs because it is a task that can be objectively assessed. As explained in the post below, 'Programming has a unique advantage: through 'self-play', it can achieve superhuman data scaling. Models can write code and then run it, or write code, write tests, and check for self-consistency.'
Limit the negative effects of hallucinations—current models are very powerful but far from perfect. Agents cannot be given complete freedom to execute operations.
Promote non-speculative application scenarios—such as allowing you to buy tickets through @xpticket, optimize yields for stablecoin portfolios, or order food on DoorDash.
Currently, while LLMs are far from perfect in code writing and have some significant shortcomings (e.g., they perform poorly at identifying vulnerabilities), tools like Github Copilot and AI-native code editor Cursor have fundamentally transformed software development (even changing the way companies recruit talent). Given the expected rapid progress, these models are likely to revolutionize software development. We aim to leverage this progress to enhance Solana developers' productivity by an order of magnitude.
However, there are currently some challenges hindering LLM's performance in understanding Solana:
There is not enough high-quality raw data available for LLMs to train on;
Lack of sufficient verified builds;
There is a lack of sufficient high-value information exchange in places like Stack Overflow;
The rapid development of Solana infrastructure means that even code written six months ago may not fully meet current needs;
There is no way to assess the model's understanding of Solana.
What we hope to see
Help us publish better Solana data on the internet!
More teams to release verified builds.
We hope to see more people actively engage with Stack Exchange within the ecosystem, asking good questions and providing high-quality answers;
Create high-quality benchmarks to evaluate LLM's understanding of Solana (RFP coming soon);
Create higher-scoring fine-tuned versions of LLMs in the benchmarks mentioned above, and more importantly, accelerate the work of Solana developers. Once we have high-quality benchmarks, we may offer rewards for the first model to reach the benchmark score—stay tuned.
The ultimate achievement here will be a high-quality, differentiated Solana validator client created entirely by AI.
3) Support an open and decentralized AI tech stack
Why are we focused on this? It is currently unclear how power in the AI space will balance between open-source and closed-source AI in the long term. There are strong arguments for why closed-source entities will maintain the technological frontier and capture most of the value from foundational models. The simplest expectation now is that the status quo will continue—large companies like OpenAI and Anthropic will push the technological frontier, while open-source models will quickly follow and ultimately possess uniquely powerful fine-tuned versions for certain use cases. We hope Solana can closely interface and support the open-source AI ecosystem. Specifically, this means facilitating access to the following: data for training, computational power for training and inference, model weights, and the ability to verify model outputs. We believe this is important for several specific reasons:
A. Open-source models help accelerate debugging and innovation in model development. The open-source community's ability to quickly refine and fine-tune open-source models like Llama demonstrates how effectively the community can supplement the efforts of large AI companies in advancing the frontier of AI capabilities (even Google's researchers pointed out last year that 'we have no moat regarding open-source; OpenAI doesn’t either'). We believe that a thriving open-source AI tech stack is crucial for accelerating the pace of progress in this field.
B. Provide an exit for those who may be forced to use untrusted AI (e.g., state-sanctioned AI). AI could now be the most powerful tool in the arsenal of dictators or authoritarian regimes. State-sanctioned models provide a state-approved version of the truth and become a massive control mechanism. Highly authoritarian regimes may also have better models because they are willing to overlook citizens' privacy to train their AI. The question of when AI is used as a control tool is not if it will happen but when. We hope to support an open-source AI tech stack to prepare for this possibility.
Solana has already become a home for many projects supporting open-source AI tech stacks:
Grass and Synesis One are facilitating data collection;
@kuzco_xyz, @rendernetwork, @ionet, @theblessnetwork, @nosana_ai, etc. are providing a wealth of decentralized computing resources.
Teams like @NousResearch and @PrimeIntellect are working to develop frameworks that enable decentralized training (see below).
What we hope to see is more product development at all levels of the open-source AI tech stack:
Decentralized data collection, such as @getgrass_io, @usedatahive, @synesis_one
On-chain identity verification: including protocols that allow wallets to prove they are human identities, as well as protocols to verify AI API responses so consumers can confirm they are interacting with LLMs
Decentralized training: for example, @exolabs, @NousResearch, and @PrimeIntellect
Intellectual property infrastructure: enabling AI to license (and pay for) the content it utilizes
