Amadeus Protocol ($AMA)

Amadeus Protocol ($AMA): A Live Confidential L1 for Agentic Finance, Still Pre-TGE

Amadeus Protocol is a high-performance Layer-1 blockchain designed as confidential settlement infrastructure for AI agents, particularly in finance. It combines Useful Proof of Work (UPoW—miners perform real matrix multiplications / AI-relevant compute rather than empty hashes), Trusted Execution Environments (TEEs) for private execution, deterministic agent runtimes (same inputs produce the same outputs), a Nova AI orchestration layer, WASM contracts, sub-second finality (~0.5s), and native payment rails (including x402-style). The network has been live on mainnet since March 2025. 

As of late October 2026 the chain reports hundreds of millions of transactions, tens to hundreds of thousands of daily active users (figures vary by update), tens of thousands of agents, and integrations across Base, Solana, Robinhood Chain and others. AMA Hub and related tools allow users to compose, deploy, and run agents. The native token ($AMA) has no public market price yet. Official channels have described the TGE as “coming very soon” (most recently around 20 October 2026) while emphasizing ongoing activity and lockups. There is no disclosed traditional VC or institutional raise; the project positions itself as a fair-launch model.  Primary Sources

• Official website and product pages: ama.one, including participation, institutional, roadmap, and about sections.

• Litepaper v1.0 (January 2026): available via the site (ama.one/docs/litepaper.pdf). Covers architecture, tokenomics phases, acquisition of Bitte.ai, and team acknowledgements.

• Shenron Upgrade tokenomics post (Medium, December 2025): detailed emission transition.

• Official X accounts: @ama_protocol and @AMA_Fdn (Foundation).

• GitHub (amadeusprotocol), docs, explorer, and validator program pages (amafoundation.net/validator).

• Public team profiles on the about page, LinkedIn, and F6S. These are the core primary materials; secondary airdrop trackers and summaries should be treated as secondary. Tokenomics (Published vs. Unpublished) Hard cap: 1,000,000,000 AMA. No inflation mechanism beyond the scheduled emission curve; no emergency mint.  Distribution model: Litepaper states a fair, miner-only launch—100% of supply emitted through UPoW, no VC allocation, no premine, no team lockups. Circulating supply was roughly 450 million AMA as of December 2025.  Emission schedule (Shenron Upgrade and later refinements):

• Early phases used a linear curve (high issuance for bootstrapping, e.g., hundreds of thousands of AMA per epoch).

• At epoch 750: transition to logarithmic decay. Total issuance drops ~65% (example figures from the Medium post: ~779k → ~271k AMA per epoch). Mining fees simultaneously decline (from high levels toward 25%), so net miner rewards ultimately rise relative to the prior schedule.

• Post-transition: emissions split between yield vaults and uPoW solvers, with a protocol/treasury allocation (up to ~25% in some descriptions), under the fixed 1B cap. Logarithmic decay continues toward the hard cap over an estimated 15–30 years (approaching full supply around year 20 in one projection).

• Epoch length is on the order of ~13.9 hours in current descriptions; rewards are tied to measurable compute (AMAFLOPS / tiles) with a floor so emission does not go to zero.  Utility: Transaction/execution costs (sub-cent, capped), agent payments, miner/validator incentives, yield vaults, and planned governance. Yield vaults allow locking (3/6/12-month terms with nominal early APYs in the 5–20% range, plus bonuses for early 12-month locks) that also back validators. Unpublished or incomplete fields (as of late October 2026): Exact TGE date and listing details; precise PRIME points → $AMA conversion rate or airdrop size (activity on AMA Hub is incentivized and “counts toward final reward allocation,” but parameters remain undisclosed); full current circulating supply and exact live emission rate; complete long-term vesting or unlock schedules for any non-emission holdings; final staking rewards pool payout mechanics and duration (Foundation materials reference a 60M AMA staking pool from TGE and post-TGE mining allocations). Team lockup details appear in operational posts rather than the litepaper (see below). Dilution and Unlock Analysis The model avoids classic VC cliff unlocks and a large premine. Remaining supply is emitted gradually via the decaying curve, so post-TGE dilution from new issuance is expected to be modest and predictable rather than front-loaded. However, several factors affect effective float:

• Substantial circulating supply already existed by end-2025 (~450M). Early miners and participants therefore hold a large share.

• Official posts describe team/company wallets locking ~400M AMA (40% of supply) into yield vaults pre-TGE, with additional circulation expected to enter locks (~500M total locked in one description). One clarification stated company-owned tokens would be locked in vaults that earn no staking rewards. Vaults have fixed terms and unlock windows (e.g., 21-day unlock process after maturity). This reduces immediate sell pressure but creates future unlock events as terms expire.

• Founding validator program and treasury can provide upfront stake for selected partners; these arrangements have agreed terms.

• Ongoing emissions continue (decaying) and are split among solvers, vaults, and treasury. High early issuance means a large portion of the 1B is already or will soon be distributed to compute contributors. Net effect: lower structural dilution risk than many VC-backed launches, but meaningful floating supply already exists and periodic unlocks from vaults will occur. Demand for real agent execution, fees, and staking must absorb emissions and unlocks for price stability. Exact modeling requires current epoch, total stake, and compute participation (available on the explorer). Team Verification Core publicly listed team (from ama.one/about and supporting materials):

• Ivan Svirid — Co-Founder. Distributed systems engineer (13+ years), background in high-concurrency systems (Erlang/Elixir, Rust), former NEAR ecosystem contributor. Appears in technical interviews and streams.

• Leslie Kivit — Co-Founder (joined publicly around early 2026). Operator with 15–20+ years scaling products; prior roles at Qubic, Xapo Bank, Meta (Oculus/Libra), Booking.com, Rocket Internet. Leads GTM, ecosystem, and operations. LinkedIn and F6S profiles align with official announcements.

• Valentyn Faychuk — Lead Backend Engineer. Prior experience in decentralized systems.

• Additional contributors and advisors listed in the litepaper acknowledgements; broader team size described as roughly 10–50 people. Company linked to Delaware and Lisbon/Zurich references; AMA Foundation handles certain non-profit / decentralization aspects. Public signals include consistent technical shipping (node upgrades, vaults, TEE work, Hub), Bitte.ai acquisition (December 2025), and open GitHub activity. No major anonymous team red flags in primary materials, though full backgrounds and any undisclosed advisors should be independently verified. Key Risks

• Pre-TGE valuation and liquidity risk: No market price; “coming soon” timelines have slipped relative to earlier mid-Q3 2026 targets in some updates. Price discovery at TGE will be highly uncertain.

• Utility and adoption: Token value depends on sustained demand for confidential agent execution, fees, and compute. Metrics (transactions, agents, EAV claims) are positive but early; competition from other AI-native or agent-focused chains and established L1s integrating AI is intense.

• Technical and privacy risks: TEE reliability, deterministic guarantees under complex agent workloads, Nova AI performance, and bridge/security surface area. Enterprise compliance claims (SOC 2, ISO, HIPAA, GDPR, MiCA references) require ongoing audits and real deployments.

• Token supply dynamics: Large early circulating supply + vault unlocks + continued (though decaying) emissions create potential sell pressure if demand does not materialize. Fair-launch narrative coexists with significant team/company lockups of emissions.

• Centralization and governance transition: Early designated validators are being phased out toward vault-backed and top-solver sets; full decentralization is still in progress.

• Regulatory and operational: Agentic finance, private strategies, and cross-chain activity face evolving rules. Execution risk on roadmap items (Stellaris trading vertical, on-prem enterprise, full marketplace, inference serving) remains.

• Information gaps: Incomplete public tokenomics parameters and points conversion increase uncertainty for participants. Holdings / Incentives Disclosure I am an AI research assistant with no personal holdings, positions, tokens, or financial incentives in Amadeus Protocol or any other project. This analysis is based solely on publicly available primary sources as of 26 October 2026 and is not financial, investment, or legal advice. Crypto assets, especially pre-TGE projects, involve substantial risk of total loss. Readers should verify all claims directly against official channels and consult qualified professionals.

Amadeus stands out among pre-TGE projects for having a live mainnet with real (if early) agent activity and a compute-backed emission model rather than a pure points farm. The combination of fair-launch claims, decaying emissions, and heavy pre-TGE locking is structurally interesting, but the ultimate test will be whether confidential agentic finance generates durable demand for $AMA once the token becomes liquid. Monitor official channels, the explorer, and primary docs for updates.