Starknet to bring on AI agents that can carry out on-chain activity

As a seasoned crypto investor with a strong background in technology and finance, I’m excited about the potential of Ethereum layer-2 blockchain Starknet integrating AI agents for autonomous on-chain activities. The ability to optimize yield and manage portfolios using AI is a game-changer that could significantly enhance returns while reducing the need for constant monitoring. Giza, a pioneering firm in this space, is leading the charge with $3 million in funding from reputable investors like CoinFund.


Starknet, a layer-2 blockchain solution on Ethereum, is set to introduce AI agents. These intelligent agents will be capable of independently executing on-chain tasks for users, including enhancing returns by optimizing yields and readjusting investment portfolios.

Giza, a company focused on integrating artificial intelligence (AI) into blockchains, secured $3 million in a pre-seed funding round headed by CoinFund in July 2023. The team at Giza is taking the lead in this initiative. According to co-founder Renç Korzay, AI agents will be available on Starknet by the end of June, with additional blockchains set to follow suit.

According to Korzy, agents have the capability to assume control over four distinct functions. The initial activity involves yield aggregation, which essentially entails identifying where the maximum return is obtainable and allocating funds accordingly across various yield-generating channels.

In the future, our AI agents will not only interact with users but will additionally function as “robo-advisors,” determining optimal investment portfolios based on individual risk preferences.

As a member of the Giza team in the beta stage, I will be responsible for reviewing and endorsing agent strategies to maintain security. In the long run, through the Giza Protocol, agents will gain complete permissionlessness, offering both cryptographic and cryptoeconomic safeguards that ensure a secure and trust-minimized interaction between developers and users.

In addition, they have the capability to independently supply liquidity on decentralized trading platforms and place leveraged wagers on pricing trends, modifying their cryptocurrency assets according to the predefined user-set strategy for anticipating price fluctuations.

As an analyst, I would interpret this statement as follows: I have learned from a May 9 press release obtained by CryptoMoon that yield-optimizing AI agents will lead the deployment process, while other types will follow suit.

This method is described as “affordable and computationally light,” and it optimizes returns by executing buy or sell orders for assets to maintain their proportions in a portfolio on lending platforms, granting it the authority to execute transactions directly.

“Giza pointed out that since every account on Starknet functions as a smart contract due to its account abstraction capabilities, it’s possible to authenticate verifications right within the contract itself.”

Starknet to bring on AI agents that can carry out on-chain activity

AI models are constructed utilizing the Starknet programming language, Cairo, from StarkWare, and they authenticate intricate calculations through Zero-Knowledge Scalable Transparent Argument of Knowledge (zk-STARKs) and zero-knowledge proofs (ZKPs) developed by the same firm.

ZK-proofs allow for information to be shared and authenticated while maintaining user privacy.

As a crypto investor, I can tell you that according to Giza’s co-founder Cem Dagdelen, we should anticipate that agents will serve as the new interface for Web3, shielding us from the intricacies and potential risks associated with smart contracts.

Yet, these agents aren’t restricted to only crypto and blockchain. As per Giza’s statement, they can assist “highly regulated industries” in reducing compliance expenses by employing models to verify “the authenticity of their results.”

At a recent blockchain conference, Ava Labs founder Emin Gün Sirer expressed his belief that the development of artificial intelligence (AI) in smart contract coding is a prospect between five to ten years in the future. Once realized, this advancement could potentially attract billions of new users to the blockchain platform.

Vitalik Buterin, a co-founder of Ethereum, has advised caution for developers integrating AI into blockchains. He believes that AI can aid humans in interpreting on-chain activities, but it should not be responsible for implementing smart contract regulations.

Read More

2024-05-09 13:48