Join our daily and weekly newsletter for the latest updates and exclusive content on industry-leading AI coverage. learn more
We’ve heard a lot about AI agents. This is a tool equipped with a generator AI model that allows you to perform actions without the supervision or intervention of many human beings.
But they still remain a novel curiosity for most people, and as far as we know, very few people trust AI agents to buy or enter contracts on their behalf.
GenLayerStartups from Stealth believe there is technology that provides the “trust” component that is lacking in the AI agent economy.
The Genlayer idea is a blockchain-powered infrastructure that allows AI agents to draft contracts, resolve payments, and execute contracts autonomously.
Last fall, the company announced that it had risen. $7.5 million This vision is realized from notable investors including Arthur Hayes (Maelstrom), Arrington Capital and North Island Ventures.
How to make AI agents trust people – and each other
AI agents have already proven their ability to analyze data, trade and manage assets, but there are fundamental issues. They essentially don’t trust each other. Unlike humans, AI agents should not be afraid of lawsuits or reputational damage.
Albert Castellana, CEO of Yeagerai (Genlayer’s company), considers this a serious flaw in AI development today.
“How do they trust each other, even in a situation where there are agents who can do business with themselves?” Castellana said in a recent interview with Venture Beat. “AIs don’t sleep, AI works worldwide, AI can’t go to prison. There’s a big problem with the legal system dealing with this kind of situation.”
Traditional smart contracts where power blockchain-based transactions are rigid for AI-driven commerce.
They cannot process unstructured data, understand complex languages, or adapt to real-world changes.
GenLayer wants to upgrade its smart contracts to “intelligent contracts.” This is a more flexible, AI-driven contract that is much like a human contract.
Unlike traditional blockchains where external oracles need to access off-chain data, GenLayer integrates AI directly at the protocol level. Intelligent contracts can handle live web data, natural language input and natively get reasons for complex, real-world conditions. All of this is done independently of third-party services.
“Blockchain allows for self-forced contracts, but there are restrictions,” explained Castellana. “They can’t connect to the outside world. They can’t understand unstructured data. But AI needs contracts that resemble human contracts. They’re fast, cheap and adaptive.”
GenLayer says, “An optimistic democracy, an AI-driven consensus model (each of which is a different, large language model (LLM)) resolves by voting on whether the AI-generated contract or decision is valid. This ensures that a single AI model is controlled and cannot be prevented from manipulating.
“We created a blockchain where validators can reach consensus even if they get different responses from AI and the Internet,” Castellana said. “It’s basically a court system for the future of commercial.”
How the Genlayer approach works
GenLayer acts at its core as an AI-Native Trust Layer. This is an independent system that ensures that AI agents operate quite well in financial transactions, contract execution, and dispute resolution.
The main features are:
- Intelligent contract:AI-driven contracts that process natural language and access live web data.
- AI-driven decision making: A consensus model in which multiple AI models vote for results to ensure reliability.
- Optimistic democracy: A blockchain-based governance model that prevents AI manipulation using decentralized decision-making.
- On-chain and off-chain interoperability: Ability to connect smart contracts to actual data and internet sources.
- zksync integration: Scalability, low cost, Ethereum-level security.
At the heart of Genlayer is “optimistic democracy,” a delegated, delegated proof (DPOS) model that integrates AI directly into blockchain verification. Instead of relying on deterministic logic, variators connect to LLM to process natural language, interpret data, and perform complex decisions in the chain.
When a transaction is sent:
- The reader validator processes the request and proposes results.
- A set of validators validates the leader’s proposal and independently recalculates the transaction.
- If the majority agree, the transaction is finalized. Otherwise, a new leader is selected and the process is repeated.
This mechanism prevents manipulation and ensures that AI-generated decisions are supported by consensus rather than by a single entity’s decision.
Inspired by Condorce’s ju judge theorem – 18th century mathematics and political science theory that ju apprentices are likely to reach the right decision with more participants – The system aggregates AI output across multiple validation factors to ensure fairness and reliability of non-deterministic tasks such as interpreting legal contracts, validating supply chain data, or setting up dynamic pretreat models.
This approach is explained in a white paper published by three co-founders of Genlayer, Castellana, José María Lago and Edgars Nemše. It is embedded below.
Why Genlayer thinks that moment has arrived?
The competition to create an autonomous AI business is accelerating. Companies like Openai have deployed AI agents that can work independently, but still rely on slow, human-driven legal and financial systems.
“AI will not wait for a lawyer,” Castellana emphasized. “If you want to participate in the economy with AI, you need infrastructure that suits your speed.”
Other startups are working on AI-Agent transactions such as Skyfire and Pin AI, but Genlayer takes a different approach. Instead of focusing on building the AI agents themselves, GenLayer is creating a trust layer that enables trans.
“We have 100 startups working on AI agents,” says Castellana. “But trust requires a third party. We build that third party – the infrastructure that enables AI commerce.”
To encourage verification devices and to cover the costs of carrying out intelligent contracts, GenLayer introduces a native gas token called GEN. Users pay trading fees on GEN and are then distributed to the validator as a reward for the service.
This approach ensures that AI-driven transactions remain fast, low cost and self-supported. Additionally, Genlayer’s token-based staking model adjusts the incentives by rewarding honest validators and fines bad actors through thrashing mechanisms.
What’s next?
Genlayer’s testnet is launching an early project focused on:
- AI-RUN Supply Chain It negotiates logistics autonomously.
- AI-driven influencer marketing Payment is performance-based.
- Decentralized Finance Applications This uses AI to execute transactions.
- Autonomous, Decentralized Autonomous Organization (DAO) Adjust governance based on real-time community feedback.
- AI-equipped insurance and lending Adapt the term based on external risk factors.
The company will showcase its technology Ethereum Community Conference (ETHCC8) It will be held in Cannes, France from June 30th to July 3rd.
The AI economy is here, but without trust, it won’t expand. Genlayer bets that AI-driven contracts implemented on the blockchain will serve as the foundation for a trillion dollar machine-driven market.
“We are building a legal framework for global AI commerce,” Castellana said. “This is to enable AI to work together with confidence, not only faster.”
