Table of Contents
Opinion
Web3 AI space has fallen into a mesh. Somewhere, “decentralized AI” became synonymous with “blockchain AI”. This false equivalence actively damages innovation.
Excellent decentralized AI projects explore meselves into blockchain framuwers, not because it makes technical understanding, rather because the web 3 is the only way to funding, expertise, and communities. Blockchain-first mentality is not doing what AI may be decentralized; This cannibal.
Web3 is not blockchain
Web3 emerged from ideal cyperpunk: reliableness, permission, censorship resistance and user ownership. The industry has forgotten a significant difference: web 3 philosophy difference from blockchain technology. Bittorrant is a web 3. Tor is web 3. IPFS is web 3. And now that the web 3 is at the center of publicity, many people are often ill.
Jump into any web3 AI space, and you will see reiterating the same pattern: Fantastic teams built distributed learning networks, Peer-to-Pier (P2P) AI marketplace and distributed training systems, explaining all strangely why they open a token or how they open a token.
As a protest, considering learning, where many nodes collaborate to train a shared AI model that keeps their raw data private. This is a prominent example of decentralized AI: no token request is made.
This is not to say that blockchain is never useful. Onchain can simplify payment between settlement agents, apply cryptographic evidence reputation systems, and can align encouragement in token collaborative training. But they are special tools, not a size-fit-al-selection. For many decentralized-AI problems, blockchain overhead only connects lathency, complexity and cost.
A game of incontions
Why do these decisions make these decisions? How the web3 ecosystem has developed is the answer. Projects that do not integrate blockchain are “web 3” scenes and thus the web 3 funding pipelines, specialist networks or community resources cannot reach. In his thesis, Venture Fund with “Web 3” created an investment criteria around blockchain integration. Web3 AI space consider non-blockchen projects out of the realm.
These intentions motivate blockchain to adopt blockchain not due to product reasons but for ecosystem acce. In other words, they are taking architecture decisions based on the requirements of wealth raising money thanks. The game is not with the play, but it means that many opportunities for the actual (and profitable) applications of decentralized AI are being overworked.
The industry should recognize that three separate conception is artificially bundled together.
Connected: Can Blockchain AI supply guards to keep AI?
Decentralized AI includes computing, federated learning, P2P network and edge computing, which is not a blockchain infrast in any way.
Crypto-ecclested AI include token tokens intentions, cryptographic proof, digital asset management and leggitimat in cases that can be applied using blockchain.
The web3 AI represents the user ownership, permissionless innovation and community rule, which can be taken through many technical approaches.
These concepts can work beautifully together, but they do not need it. A federated learning network can use a cryptographic privacy guarantee with a blockchain. A distributed AI marketplace can perform without-based verification without a smart contract. Intentitive systems can work through Kansas mechanisms that do not require overhead of the entire blockchain infrastructure.
Decentralized AI requires a toolbox
Recognized AI requires true innovation requires technical pluralism, with blockchain being a toolkit, not religious requirement. The most successful projects of the next decade would be that their specific challenges choose the right generation, not for those who correspond to experts in the currency ecosystem.
Web 3 funding and community support must develop to embrace non-blockchen decentralization. Venture funds can get enough retired on aligned with decentralized and web 3, even if the funding models are not tokens-based.
The communities should celebrate the innovation -free innovation regardless of their technical substrate. For examples, many decentralized AI ecosystems exist beyond crypto, both non-profit and profit-benefit. Prime Intelligence has trained the big language model on a scale that protects decentralization. Massachusetts is creating a decentralized international of Nanda agents of the Institute of Technology. Laion AI is democratizing research.
These systems receive real decentralization; They do not carry a blockchain badge and are invisible to the web 3 community. Even more traditional web3 AI in space, how, there are positive signs from provisions that use blockchain when it makes sense.
Uses a series to manage stakes for models developed by the Numerai community, which gives the best performing ones to the river. Torus network tokens transparently distributes tokens to agents that contribute the most to their long -term girls that capture the network value in tokens. The token-based payment of the render network means that any, anywhere, can provide calculations to your render form. All these applications are already already.
The current blockchain-Foster decentralized innovation in AI is properly interrupted when it requires the most. As the AI systems become more powerful and centralized, decentralized options are in dire need. But they will be obtained if the ecosystem forces everything through blockchain bottlenecks. Today, projects that shed this disabled mindset will dominate the ecosystem tomorrow.
Web3 AI withstand an option: Continue cannibalizing decentralized AI with blockchain requirements or free it to achieve your full potential. Technology is ready. The question is that the ecosystem is ready to develop and who can do this change to capitalize on.
Opinion by: Samuel Marro, PhD student at Machine Learning at Oxford University.
This article is for genealogy information purposes and is not intention and should not be taken as legal or investment advice. The idea, however, expressed opinions here alone of the author and not necessarily reflected or represented the ideas and ideas of the coinletgraph.
