Amadeus Protocol

Amadeus Protocol is a Layer 1 blockchain designed from the ground up to create, deploy, and monetize autonomous AI agents. Rather than adapting an existing general-purpose chain, Amadeus builds its consensus mechanism and runtime around the specific requirements of agent workloads: verifiable computation, on-chain memory, coordinated multi-agent execution, and real-time settlement with sub-second finality.

The Problem: Agents Need Purpose-Built Infrastructure

Autonomous AI agents are increasingly being trusted to handle real capital, coordinate with other agents, and make consequential decisions without human intervention. Yet the infrastructure available to them — both traditional cloud compute and existing blockchain networks — was not designed with agent-specific requirements in mind. Cloud execution is opaque and unverifiable; general-purpose chains are slow and lack the native memory and coordination primitives agents need. Amadeus Protocol was founded in early 2025 to fill this gap, positioning itself as the settlement layer where agent activity becomes auditable, coordinated, and economically self-sustaining.

Useful Proof of Work: Mining That Trains AI

The most distinctive technical choice in Amadeus is its consensus mechanism: Useful Proof of Work (uPoW). Traditional proof-of-work networks direct enormous computational energy toward solving arbitrary hash puzzles that serve no purpose beyond securing the ledger. Amadeus replaces this with real Tensorcore matrix computations — the same mathematical operations used to train and run machine learning models. Validators earn block rewards and execution fees in $AMA while simultaneously performing work that advances the network's AI capabilities. The protocol describes this as creating a feedback loop where increased mining directly enhances the intelligence infrastructure available to agents running on the network.

BLS-based validation is used for consensus participation and block finalization, while the uPoW workload requirement ensures that the computational resources securing the chain are intrinsically productive rather than wasteful.

Nova AI: No-Code Agent Deployment

The Nova AI compiler is Amadeus's developer-facing layer for creating agents. Users describe an agent's intended behavior in plain language, and Nova compiles that description into verified, test-driven code — no programming expertise required. Once deployed, agents operate directly on-chain with built-in persistent memory and coordination logic. All modifications to an agent are recorded on-chain for transparency, and the execution model is deterministic: identical inputs always produce identical outputs, making agent behavior auditable after the fact.

Agents in Amadeus can also be composed into teams using drag-and-drop swarm coordination modules, with integrated oracle data feeds enabling agents to consume real-world information such as prices, news, or web data. Usage fees generated by a deployed agent flow back to the creator, creating an economic model where agent builders capture value from the activity their creations generate.

Privacy and Verifiability

Agent operations requiring privacy run inside Trusted Execution Environments (TEEs), which protect sensitive inputs and intermediate states while still producing verifiable attestations of correct execution. For broader computational verification, Amadeus has established a partnership with zkVerify, a zero-knowledge verification layer developed by Horizon Labs. The planned integration creates a proof settlement bridge between the two networks, enabling AI training and inference workloads to receive zk-proof validation. This supports enterprise compliance use cases including provenance attestation, geofencing, and licensing verification, and enables cross-chain interoperability with EVM-compatible networks through KZG-based proofs. Phase 1 covers basic uPoW job validation; Phase 2 scales to comprehensive AI computation verification.

Storage: Arweave and Crust Network

On-chain storage of large AI artifacts — model weights, training datasets, agent logs, audit records — would be prohibitively expensive. Amadeus offloads this data to Arweave's permanent decentralized storage network, retaining only lightweight on-chain hashes that serve as verifiable audit trails. The implementation uses AR.IO's Turbo bundling solution for high-throughput uploads and Rust-based Arweave crates for cryptographic signing. Crust Network provides additional decentralized storage capacity within the partner ecosystem.

Network Participants

The Amadeus ecosystem operates through three interdependent participant classes. Builders create agents and applications that generate demand for compute. Validators supply that compute by running nodes that perform uPoW matrix computations, earning $AMA in return. Partners extend the ecosystem with external capabilities: storage infrastructure (Arweave, Crust Network), verification services (zkVerify), data feeds, development tooling, and liquidity support through market makers, exchanges, and launchpads.

$AMA Token

The $AMA token is the native utility token of the Amadeus network, used to pay transaction fees and network operation costs. Total supply is fixed at 1 billion $AMA, with all tokens distributed through mining rather than a pre-mine or venture capital allocation. The protocol describes this as a fair launch: no tokens were minted to insiders at inception. A 100 million $AMA community treasury funds ecosystem grants targeting mining infrastructure, SDKs, and AI tooling for builders. $AMA is characterized in official documentation as a utility token and explicitly not as a security or investment vehicle.

Team and History

Amadeus Protocol was founded in March 2025. Ivan Svirid serves as founder and infrastructure lead, bringing over thirteen years of experience building massively concurrent distributed systems in Erlang and Elixir. The broader team includes engineers and technical leaders from major technology companies and multi-billion-dollar crypto projects, with combined experience spanning blockchain infrastructure, AI, and distributed systems.

No third-party security audit of the Amadeus Protocol codebase has been disclosed in publicly available documentation.

Roadmap

The Q4 2025 roadmap targeted the Nova Runtime deployment — activating the AI agent execution capabilities — alongside a first centralized exchange listing to expand $AMA liquidity. The 2026 roadmap focuses on enterprise adoption, with a GPU marketplace planned to allow participants to supply and access specialized compute, and an ecosystem grants program to accelerate developer onboarding. The Arweave storage integration is targeted for full deployment in Q1 2026, while the zkVerify zk-proof bridge is planned to complete across two phases spanning Q4 2025 through Q2 2026.

Context in the AI Agent Landscape

Amadeus enters the market as autonomous AI agents move from experimental to production deployments. The protocol is positioned as infrastructure-layer rather than application-layer: the settlement and execution substrate that agent-native applications are built on, rather than a specific agent product. Its combination of uPoW consensus, no-code deployment via Nova AI, and verifiable on-chain execution differentiates it from both general-purpose Layer 1 chains and off-chain AI compute networks. The project claims 16,000 active agents already operational on the network as of its public launch communications.