One of my Students during our tech session yesterday asked if there would be anything bigger than artificial intelligence (AI) in terms of technological advancement. I didn't see that question coming and did not know what inspired him, but I told them that the wake of artificial intelligence is opening a whole new chapter for advances in tech since AI can help undertake complex jobs, hence, research and brainstorming would get a lot easier.
After our session, I had to look up some new deliveries in AI so I have what to share with them next Tuesday and I discovered there is a new AI assistant that has gone live with an ambitious goal of becoming the world’s largest open-source artificial intelligence model and seeking to reach a capacity of 3.5 times bigger than Meta’s current open-source Llama model. "Could that be a joke?" I asked, but then, NEAR protocol is at it!
Near Protocol
Behold NEAR.AI
During the opening day of the Redacted conference held in Bangkok, Near Protocol unveiled their record-breaking, 1.4-trillion-parameter open-source AI model. This AI model would be the world’s largest open-source artificial intelligence model, expected to be 3.5 times larger than Meta’s LLaMA, and aims to harness decentralized, crowdsourced development through Near’s AI Research hub.
According to Near co-founder, Illia Polusukhin, speaking at Redacted in Bangkok, the Near AI Model would feature thousands of contributors collaboratively building this model, starting with a smaller, 500-million-parameter version. The model’s development is to be spearheaded by Near’s newly launched AI Research Lab. The lab will serve as a collaborative space for developers, researchers, and enthusiasts from across the globe. Near co-founder Illia Polosukhin also emphasized the importance of decentralizing AI to prevent monopolization and maintain Web3’s core principles.
Decentralized Funding
One of the project’s most innovative aspects lies in its approach to funding and incentivization. Near Protocol, a prominent player in the blockchain industry, plans to leverage blockchain technology to crowdsource both financial resources and intellectual input. The Near protocol seeks to leveraging token sales to fund the model’s $160 million development cost. Through the tokenized incentives, Near will offer compensation to contributors, encouraging the participation of thousands of AI enthusiasts and developers worldwide. This model not only democratizes the research process but also aims to make AI innovation accessible beyond traditional research labs.
Prioritizing Privacy and Decentralization
Near’s AI model aims to stand apart by placing a strong emphasis on privacy and decentralized governance. Many current AI models are controlled by centralized entities, raising concerns about monopolization, data privacy, and biases. Illia Polosukhin, co-founder of Near Protocol, emphasized that decentralizing AI is essential to prevent monopolistic control and to ensure that the model serves the global community rather than a select few. To address privacy concerns, Near plans to implement advanced encryption technologies to safeguard user data. These measures are aimed at ensuring data is processed in secure environments without compromising individual privacy.
The Web3 AI
The operations of the NEAR.AI project is to be rooted in the principles with Web3, aiming to make artificial intelligence more accessible and interoperable with blockchain technologies. Near Protocol believes that the convergence of AI and blockchain will open new avenues for both industries, allowing for decentralized applications that are more intelligent and adaptable. Near’s model could support applications that bridge the gap between user data and Web3 infrastructure, offering insights and functionalities that are currently unattainable in siloed environments.
One of the AI model’s anticipated benefits is its utility for decentralized applications (dApps) and smart contracts. Moreover, the open-source nature of the model will allow developers to experiment and build on it, fostering innovation across multiple sectors.
Near AI intents
This leap by Near protocol does not only surpass existing models in size and scope but also establish new standards for open-source AI within the Web3 space.
Trust me, our conversation next week will be robust as we now know that bigger AI developments and advancements are coming.