NEAR Rabbit Hole Ep. 3: The User-Owned, Open-Source Network for the AI Era

Captions as YouTube publishes them, usually machine-generated, so product and people names can be misheard.

1. 0:00

   Data ownership, code verifiability, and decentralization have never been more important than in the AI era. And Near is successfully tackling each of these things. Let's talk about it. Welcome to the Near Rabbit Hole content series that will guide you down this exciting rabbit hole that is the Near Protocol, exploring all its facets. No worries if you can't digest it all at once. It's part of the process. Feel free to ask

2. 0:23

   questions, correct me, like, share, follow, and subscribe. This series is sponsored by Near Dev Hub, part of the public goods for near protocol campaign. Yet everything here reflects my genuine thoughts and opinions limited to my knowledge of these things. This is episode 3 and you can watch previous episodes on X at vinballr and on the free publication the coding.substack.com.

3. 0:45

   You can also follow near onx and visit nearehub.org for more information about the ecosystem. Data ownership, code verifiability, and decentralization have never been more important than in the AI era. And here is successfully tackling each of these things. Bitcoin was a huge success by making money user owned, verifiable, and decentralized. Ethereum was a huge success by making arbitrary data and computer programs, the smart contracts, user owned, verifiable, and decentralized.

4. 1:18

   applies the strength of these projects and leverage them, delivering a more user-friendly, more scalable, faster, and more efficient system that solves both Bitcoin and Ethereum use cases, plus the growing AI use case following the same principles, userowned money and data, verifiability, and decentralization. In episode two, we looked at the info and text techn is building to power an agentic economy, exploring among other things the sharding architecture. This is relevant

5. 1:49

   here because nonsharded blockchains need to make serious compromises when scaling. Decentralization, security, accessibility, and performance are four of the most common compromises. In episode one, we saw how near has one of the lowest fees and fashionalities addressing performance and accessibility. It is also one of the most centralized blockchains with a growing 370 validators and a Nakamoto coefficient of eight thanks to low hardware requirements for running a

6. 2:20

   node. For example, I run my chunk validator at $15 per month cost using Metapool's node studio solution. You can delegate staking to me at decoding.pool.near if you want. The near core is free libé open-source software which means anyone can verify its code, build and run it locally, propose fixes or improvements and even fork it. The network is permissionless, meaning anyone can use nearore to connect with other nodes and

7. 2:52

   actively participate in the network as an RPC node, a block producer, a chunk validator like me, or an artal node. All that without asking permission from a central authority. This episode is also not just about what Near is, but what it is enabling others to build on top of the protocol following the same premises of user ownership, verifiability, expanding it to AI. Not many people are talking about the Orwellian dystopia we could be entering in a world where most

8. 3:24

   people are using centralized unverifiable data harvesting closet source AI for many activities like reading the news, shopping, factchecking, writing code and learning. Big tech is harvesting users data linked to behavior like shopping hobbies and keyword searches, financial activity like bank accounts, payments and income and geolocation through IP. This is dangerous. Mir solve this with tools to build apps that users own their data and that also have privacy and

9. 4:02

   confidentiality primitives protecting this data from outsiders. Check the CDK developments I investigated in a thread for example. Also, a central controller can manipulate AI outputs if they can temper with the code and genesis prompt while we can't verify and audit the code independently.

10. 4:27

    This opens doors to mass manipulation both politically and consumer-wise via potential bias responses from the models. Near solves this with open-source verifiable AI related software which can also be registered on chain. Near is the cipher plank dream coming true in the AI era protecting and giving control back to the users. Let me know your thoughts in the comments.

11. 4:56

    Like, share, follow and subscribe for more. See you.