FHE Secured Restaking Layer: Scaling Security for AI & DePin Networks

Mind Network
5 min readMar 21, 2024


TLDR; FHE for Restaking and AI/Depin Networks

Restaking is a rather genius idea for both users and networks. By leveraging Fully Homomorphic Encryption (FHE), Mind Network introduces an FHE Restaking Layer built upon the restaking framework to further scale consensus security.

AI projects and systems are growing exponentially. AI compute demand is doubling every 3.5 months. By adding FHE validators to a Proof of Intelligence (POI) consensus mechanism, MindNetwork adds better capability and high security to AI systems in Web3.

Restaking: A Powerful New Trend

Restaking leverages Ethereum’s consensus layer to extend cryptoeconomic security to additional applications on the network. Ethereum is the most secure PoS chain in the world with $50B in total value locked. But the average staking yield is ~4%. Bitcoin’s current market cap is currently at $1.4T and obviously, staking is not an option. Projects like Babylon are using a self custody method and locking Bitcoin to secure proof of stake chains.

However, challenges are associated with restaking:

  • Centralization: one major concern is that restaking could result in stake centralization, leading to a few validators cheating and controlling the stake. This could lead to a potential loss of neutrality.
  • Compounded Risk: LSTs from one ecosystem (e.g. ETH) may suffer and amplify the compounded risk to become systematic for the ecosystem.
  • Data Security: operators are not equipped with data protection mechanisms, which may impact data privacy and computation fairness. It is critical for data intensive networks like AI or DePin networks.

Restaking represents a shift towards more passive, yet proactive investment strategies on Web3. It is yet another DeFi tool that can attract long term investors and builders into Web3.

FHE — the “Holy Grail of Cryptography”

Fully Homomorphic Encryption is a technique that allows you to perform computations on encrypted data. Recent advancements improved the speed of the FHE scheme by a factor of 20x, and are on track to reach 100x soon. This key performance milestone has unlocked critical blockchain and AI use cases. It is now possible to convert an AI model into an FHE equivalent, and to both train and run inference on encrypted data. Data scientists can already start building simple confidential AI applications today, and integrate more complex models such as LLMs as hardware acceleration becomes available in the near future. FHE, the “Holy Grail of Cryptography” will soon make the entire internet encrypted end-to-end via a new, FHE-powered “HTTPZ” protocol.

An FHE Restaking Layer

Mind Network is one of the first tokenized startups to implement FHE within Web3. It is designed as a FHE computation network and brings Zero Trust principles to on-chain data security. Built upon its core FHE computation network, Mind Network introduces an FHE Restaking Layer to secure AI and DePin networks with cutting edge FHE techniques. Collaborating with the best-in-class partners, Mind’s FHE restaking solution is structured with three interconnected layers:

  • Restaking Layer: we support a variety of security appetites with a range of restaking tokens. Partnering with industry leaders like EigenLayer, StakeStone, Renzo, Babylon, Ankr, and more, we bring security from best-in-class networks like Ethereum and Bitcoin through restaking tokens. We are also integrating with our strategic partners like Chainlink CCIP and Connext to offer remote restaking across different networks. Our well-received product, MindSAP (FHE + Stealth Address), is natively supported in order to protect privacy and security for onchain and cross-chain transactions.
  • FHE Security Layer: we believe the path to ensure ownership in Web3 is end-to-end encryption, which is especially critical to data intensive networks like AI or DePin networks. We introduce FHE enhanced validators to ensure end-to-end encryption for the validation and consensus computation process. This computation is fully decentralized and the owner of FHE validator could opt-in to different networks and earn their rewards. Security will be further improved by the integration with our modular fhEVM partners Fhenix and Inco.
  • Consensus Layer: Mind Network introduces a novel Proof of Intelligence (POI) consensus mechanism specifically designed for AI tasks, ensuring fair and secure reward distribution among FHE validators, which provides a robust framework for secure and transparent execution of actions within the network. Furthermore, Mind Network is launching a rollup chain with AltLayer, EigenDA, and Arbitrum Orbit, which will further enhance consensus computation with lower cost and faster performance.

Scaling Security for AI and DePin Networks

The integration of artificial intelligence into Web3 poses additional critical security challenges regarding the protection of data and models. Mind Network spearheads efforts to address these challenges through its innovative consensus mechanism.

A Proof of Intelligence (POI) consensus mechanism ensures fair and secure reward distribution among network validators, called the FHE Validator, which provides a robust framework for secure and transparent execution of actions within the network.

Mind Network’s innovative approach to bolstering Web3’s AI security is centered around the integration of an FHE-powered POI consensus mechanism, which fosters enhanced security, fair reward distribution, and efficient AI service validation processes. This integrated approach aims to drive innovation and scalability in the AI and blockchain domains, empowering users with control and privacy in a decentralized ecosystem.

FHE computations are also accelerated via GPUs. Novel AI projects, like Io.net, offer a decentralized network for distributing GPU power. We envision that Io.net’s platform will lower the barrier for AI development, making it easier for AI startups to leverage high power AI computation via GPU farms from around the world.

About Mind Network

Mind Network is the first FHE Restaking Layer for POS and AI Networks. Our framework operates as an FHE validation network, bringing secure computation and consensus to AI, DePIN, EigenLayer AVS, Bittensor Subnet, and many critical POS networks. Mind Network’s solution has found initial product-market-fit with projects such as IO.Net, MyShell, Bittensor, AIOZ, Nimble AI, ChainLink, Connext and more adopting its solution. It has achieved 650k+ active users and 3.2M+ transactions processed on its testnet.

Mind Network is backed by Binance Labs, Hashkey, Big Brain, Chainlink, Comma3, SevenX, and received an Ethereum grant to conduct research on practical implementations for FHE technology on Ethereum.

Visit Our Website and Social Media Channels:

Website | Twitter | Medium | YouTube | Github | Gitbook | Discord | Telegram



Mind Network

An FHE (Fully Homomorphic Encryption) Restaking Layer for POS and AI Networks. https://linktr.ee/mindnetwork_xyz