The Web3 and AI Newsletter 25
Hello, everyone, and warm welcome to all newly joined subscribers!
This is the Web3 + AI newsletter, where we explore the intersection of blockchain and artificial intelligence. I was away for a while attending ETHBelgrade where I listened to an amazing talk by Ora's Cathie So and had a wonderful time meeting fellow blockchain nerds. I am back now to deliver another portion of exciting Web3 + AI news!
Thank you for being here! Let's dive in!
What's New in the Web3 + AI Startup World?
Oasis Labs & Google Cloud to Protect Sensitive Data in AI
Oasis Labs, a core contributor to the private EVM layer Oasis Protocol, has introduced its privacy-preserving data analytics tool, Oasis PrivateSQL, to Google Cloud Marketplace.
PrivateSQL employs state-of-art differential privacy technology to enable the responsible use of sensitive data in AI and analytics, empowering organizations to harness the full potential of their data, derive actionable insights and make smarter decisions without compromising the privacy of individuals.
PrivateSQL makes it possible to generate privacy-preserving versions of aggregate statistical queries on a wide array of SQL databases. It is already being used to pioneer new approaches and uses of AI in sensitive and highly-regulated industries like healthcare.
The Oasis Labs Team
Atoma to Empower Sui Builders with AI
The decentralized network for verifiable AI inference, Atoma is to provide its state-of-the-art AI solutions exclusively to Sui, a Layer-1 blockchain famous for its lightning-quick and parallel transactions, and infinite horizontal scaling. Through this partnership, the Sui dApp builders will be able to integrate and leverage AI across various use cases in DeFi, gaming, agents, auditing, etc.
This collaboration will also integrate AI into Sui’s APIs and SDKs, reducing the time it takes to use certain AI-driven apps, and could soon help builders create no-code interfaces and AI-based natural language interactions.
Chainwire
NEAT Receives 1M in $NEAR to Advance AI Apps
A couple of weeks ago, NEAR proclaimed that supercharging open-source and decentralized AI will encompass the next stage of its roadmap, and it already started delivering. It partnered with its ecosystem project NEAT Protocol and provided it with a 1 million $NEAR token staking delegation. The funding is aimed at accelerating the AI applications built on NEAR.
NEAT is a rollup scaling solution built on the NEAR inscription standard, specifically designed to facilitate AI and machine-learning applications on NEAR. NEAT offers an innovative approach for rollups to be built on NEAR. With off-chain indexer nodes for execution and NEAR L1 as the data availability layer, NEAT eliminates the need for rollups to deploy smart contracts or run verification logic on the L1. This streamlined process drastically reduces transaction costs for rollups, making NEAT particularly well suited for computationally intensive applications such as AI and gaming.
NEAR Team
NEAR also published a case study exploring another AI-enhanced member of its ecosystem, namely Cosmose AI. It is building KAIKAI, an AI-powered, opt-in interactive lock screen and mobile app that provides curated content and shopping based on user preferences, and KAI-Ching, a reward system incentivizing users to engage with KAIKAI. KAI-Ching has been one of the most popular and largely used decentralized applications in the Web3 space, with over 2 million monthly active users.
NEAR Team
Ritual Releases an Infernet Library Upgrade
The modular execution layer for AI Ritual has released an upgrade of its Infernet library, a Python SDK that provides a set of tools and extendable classes for creating and deploying machine learning workflows.
[Infernet] is designed to be easy to use, and provides a consistent interface for data pre-processing, inference, and post-processing of data.
The latest version, Infernet 1.0.0, adds payments enabling nodes and verifiers to earn for providing compute and services, inbox for batching requests, and routing for unified node discovery.
Users can easily download models from anywhere- your local file system, HuggingFace and even Arweave - to run on the node. You can now also *upload* models to Arweave for storage and retrieval thanks to our Ritual Arweave library.
Infernet expands the horizon of what's possible on-chain by empowering builders with crypto-native ML workflows across a wide suite of models, primitives, and tooling.
Today Unveils AI Character Engine Dreamia
The Web3 social simulation game Today has released Dreamia, an AI-driven engine to create intelligent and interactive non-player characters (NPCs) and shape storytelling in real-time.
Dreamia is Today’s next-generation character engine. It uses AI to create intelligent, interactive NPCs that are hyper-realistic and dynamically responsive. As players engage, their actions shape the story, making every encounter in Today fluid, fun, and deeply personal. No two players have exactly the same experience which deepens the game’s authenticity, complexity and stickiness, the company said.
Jenny Jordan
Check out this great conversation with Today's co-founder Michael O'Connor for Jon Jordan's Blockchain Gaming World podcast, where he breaks down Today's strategy and vision in integrating blockchain and generative AI in gaming.
Bitcoin Miner Core Scientific Strikes AI Deal
Back in April, right after the latest Bitcoin halving, a recurring event that cuts the Bitcoin miners' rewards in half, I expressed concerns that some network operators may choose to switch to a more lucrative business like AI. This is exactly what Core Scientific, one of the largest US Bitcoin miners, is in the process of doing. The BTC mining company has struck a 12-year contract to provide resources to AI cloud provider CoreWeave for the combined amount of $3.5B. Although, fortunately, Core Scientific has agreed to lend only 18% of its total infrastructure, there is no guarantee that, at some point, it will not convert its entire data centre capacity to AI.
These developments testify to the insatiable appetite for AI computation resources the technology world is currently experiencing.
About 83% of data-center capacity under construction has already been leased in advance, with AI companies and cloud-service providers driving the demand, according to a March report by commercial real estate firm CBRE Group.
According to Bloomberg, CoreWeave even made a $1B offer to acquire Core Scientific.
Noteworthy Podcasts, Articles, Events
The Web3 and AI Funding Bulletin | May II
My Funding Bulletin covering all Web3 + AI investment news over the second part of May is out. Most of the money rounds during this period were again concentrated on AI-enhanced blockchain gaming solutions, seemingly the most appealing sector for VCs at the moment.
Albena Kostova-Nikolova
EU to Prepare for Blockchain and AI Convergence
The European Blockchain Observatory and Forum (EUBOF), an initiative of the European Commission set up to analyse the development of the blockchain space and make policy recommendations, published its latest report in May. In it, the organization suggests that the ongoing convergence between blockchain and artificial intelligence is poised to shape the blockchain landscape for the years to come, and advises that the EU should prepare for it.
One key area of convergence was data security and privacy, where the report said that blockchain could provide a secure and transparent framework for AI data, ensuring data integrity and privacy. This would be particularly important in sectors such as healthcare and finance, where data security is paramount.
Another was decentralized AI, with EUBOF suggesting that AI models and data distributed across a blockchain network could potentially reduce the risk of data monopolies and promote collaborative AI development.
James Field
Disclaimer: None of this should or could be considered financial advice. You should not take my words for granted, rather, do your own research (DYOR) and share your thoughts to create a fruitful discussion.
Member discussion