Islamic Finance Principles Assessment
Riba — Does Freysa AI involve interest?
Freysa AI's base protocol does not charge or pay interest; revenue comes from ETH query fees split between a prize pool, FAI conversion/refunds, and reported buybacks. There is no evidence of lending, bond-like fixed yields, or interest-bearing treasury instruments at the protocol level. On riba specifically, Freysa AI appears largely clean, though third-party platforms offering FAI-collateralized borrowing sit outside the base protocol's own design.
Assessment: Riba Dominant
Score: 44.1/100
Our methodology examines 10 criteria to evaluate how well Freysa AI avoids interest-based mechanisms.
Freysa's revenue is generated from escalating on-chain query fees paid in Base ETH by participants in the adversarial agent game, with roughly 70% routed to the prize pool and 15-20% converted into FAI and refunded to the payer. No sources describe this treasury holding interest-bearing instruments, issuing debt, or generating yield through lending markets. The absence of any disclosed interest-based income stream at the base-protocol level is a genuinely positive signal, though the lack of detailed treasury-management documentation leaves some residual uncertainty about how accumulated ETH is held or deployed over time.
Reported "staking" for FAI is inconsistent across sources: one describes an external platform allowing users to mint collateralized positions and borrow against staked FAI with an "APY Booster," which resembles interest-style lending but is third-party infrastructure, not Freysa's own product. Another source claims a hybrid proof-of-stake validator reward system uncorroborated by Freysa's own documentation, and a third describes agents spending FAI on their own compute ("Staking for Autonomy"), which is not a user yield product at all. Given this confusion, no fixed, interest-like native reward mechanism is confirmed to exist within Freysa's own protocol.
Gharar — How much uncertainty does Freysa AI involve?
Uncertainty here is elevated by inconsistent tokenomics disclosures, unclear team transparency at launch, and the absence of a named third-party audit. Open-source code and confirmed venture backing reduce some risk, but conflicting supply and staking descriptions leave real ambiguity. Overall, Freysa AI carries a meaningful gharar concern that Muslim investors should weigh carefully.
Assessment: Excessive Gharar (High Uncertainty)
Score: 36.1/100
Our methodology examines 15 criteria including team transparency, audit quality, and governance.
Freysa's founding link to Eternis AI (Srikar Varadaraj, Pratyush Ranjan Tiwari, Ken Li, Augustinas Malinauskas) was not proactively disclosed by the project itself, emerging instead through independent reporting despite prior informal circulation. The team later confirmed a $30M raise from Coinbase Ventures and Selini Capital, and Eric Conner, a well-known Ethereum contributor, joined in 2025, which strengthens credibility. Code repositories are open-source. Still, the initial non-disclosure of founders is a transparency shortfall that adds uncertainty for investors trying to assess accountability from launch.
No named, reputable audit firm (such as Halborn, Trail of Bits, or Neodyme, all of whom appear elsewhere in industry reporting but not in connection with Freysa) has published a specific audit of Freysa's smart contracts. The only security-related item is an automated scanner output, not a firm-authored audit with findings — this should be named plainly as an audit gap and a genuine gharar concern. Compounding this, tokenomics sources directly conflict: one describes a "100% fair launch" with burned LP, another describes private sales and multi-year vesting, leaving actual distribution terms unverified.
Maysir — Does Freysa AI involve gambling or speculation?
Freysa AI's central mechanism is a pay-to-attempt game where users pay fees trying to convince an AI to release a prize pool, with winners taking accumulated funds — a structure that functions much like a game of chance. Some genuine utility exists around agent infrastructure and governance, but the core revenue-generating activity itself carries strong maysir characteristics. This is the defining consideration for evaluating Freysa AI under Islamic finance principles.
Assessment: Maysir / Qimar (Gambling)
Score: 42.5/100
Our methodology examines 11 criteria to determine whether Freysa AI is a gambling instrument or a genuine economic tool.
Beyond the game itself, Freysa has developed an open-source Sovereign Agent Framework and API layer enabling third parties to deploy autonomous AI agents, and FAI functions as a utility token for fee payment, discounts, ecosystem access, and governance voting. This agent-infrastructure layer represents a real, productive use case distinct from pure speculation, and it is this utility — not the prize-pool game — that offers the strongest basis for arguing FAI has legitimate economic function beyond chance-based payouts.
Weighed against this utility, the base game's pay-to-play, winner-take-pool design (escalating fees up to $4,500 for a chance at treasury funds) is structurally difficult to separate from gambling, since outcomes depend on unpredictable persuasion success rather than productive output. Thin secondary-market liquidity (around $165K daily volume, ranked outside the top 400 tokens) further suggests trading activity dominated by speculation rather than utility-driven demand. The agent-framework utility is real but does not offset the gambling-like character of Freysa's flagship product.
The Full 27-Point Screening
1. Legitimacy (4 criteria)
| Criterion | Score | Analysis |
|---|
| Team Transparency | 55/100 | Team was not self-disclosed by Freysa but was identified by independent journalism, with credentialed named founders and investors. |
| Fraud & Scam Risk | 62/100 | No hack, fraud, or rug-pull is reported against Freysa itself, and LP tokens were burned, but independent security verification is thin. |
| Use Case Legitimacy | 38/100 | The documented core function is a pay-to-attempt, winner-take-prize-pool game, which is a chance-based mechanic rather than clear productive utility, even though a developer framework layer also exists. |
| Ethical Practices | 30/100 | The base protocol's own primary and original function is a fee-funded prize-pool game of chance, which is a design feature of the protocol itself, not third-party misuse. |
Summary: Freysa's team was eventually identified as credentialed AI-industry founders backed by named venture investors, though the project itself did not proactively disclose this.
2. Project Operations (9 criteria)
| Criterion | Score | Analysis |
|---|
| Core Protocol Business | 35/100 | The core protocol business model as documented is an entry-fee-funded prize pool game, a sector with gambling-like characteristics by its own design. |
| Transaction Fees | 48/100 | Fees are transparently split between the prize pool, FAI refunds to the sender, and ecosystem support, with no interest component, but they directly fund a chance-based payout structure. |
| Treasury Assets | 45/100 (low evidence) | Treasury is understood to consist mainly of accumulated ETH from fees, but no detailed breakdown of treasury holdings or interest-bearing assets is provided in the sources. |
| Revenue Model | 60/100 | Revenue is fee-based in ETH with no lending or interest component described, though it is tied to a chance-based game structure. |
| Transparency | 58/100 | Core technical components (agent code, framework docs) are open-source, but tokenomics allocation details are inconsistent across sources. |
| Governance | 42/100 | Sources assert token-holder governance over treasury and protocol decisions but provide no detail on decision-making structure, quorum, or centralisation safeguards. |
| Launch Fairness | 50/100 | One set of sources describes a "100% fair launch" with burned LP tokens, while another describes private sales and multi-year vested team/dev allocations, and the two accounts conflict. |
| Token Distribution | 38/100 | Max supply is fixed and disclosed, but specific allocation percentages are described as "opaque" in one source and contradicted by another, undermining confidence in distribution fairness. |
| Speculation/Utility Ratio | 38/100 | The project's defining and most publicized feature is a speculative, chance-based fee/prize-pool game, even though genuine developer tooling has since been layered on top. |
Summary: The base protocol centers on a fee-funded, chance-based "convince the AI" prize-pool game, now paired with an open-source agent-development framework, and its launch-fairness and distribution claims are contradicted across sources.
3. Financial Health (4 criteria)
| Criterion | Score | Analysis |
|---|
| Protocol Revenue | 62/100 | Documented revenue comes from ETH query fees rather than lending or interest, though the fee mechanism itself funds a game of chance. |
| Financial Status | 45/100 | The project secured $30M in funding from named investors, but reported trading volume is low, and no consolidated financial statement is available. |
| Interest Assessment | 68/100 | No lending or borrowing function is described at the Freysa base-protocol level itself; a related interest/collateral feature appears only on an external third-party platform. |
| Audit Quality | 8/100 | No named, reputable audit firm report on FAI/Freysa smart contracts could be located; only a generic automated scanner result was found. |
Summary: Revenue is fee-based and interest-free at the protocol level, but liquidity appears thin and no named-firm security audit of the FAI contracts could be found in these sources.
4. Token Economics (5 criteria)
| Criterion | Score | Analysis |
|---|
| Token Purpose | 48/100 | FAI is marketed with multiple stated utilities (payments, access, governance) rather than as a pure meme, though its highest-profile use case is chance-based. |
| Governance Rights | 42/100 | Holders are said to vote on treasury and protocol decisions, but no detail on scope, binding power, or process is available. |
| Rewards Distribution | 55/100 | Rewards (fee refunds, prize payouts) vary with actual participation and game outcomes rather than following a fixed schedule. |
| Speculation Controls | 25/100 | No dedicated anti-speculation mechanisms are described; the escalating fee curve is a game-design element rather than a speculation control. |
| Asset Backing | 28/100 | No described backing beyond the ETH accumulated through game fees and the token's internal utility uses. |
Summary: FAI carries several stated utilities and variable, activity-linked rewards, but its principal reward mechanism is tied to a game of chance and lacks meaningful anti-speculation design.
5. Staking Mechanism (5 criteria)
| Criterion | Score | Analysis |
|---|
| Mechanism Type | 20/100 (low evidence) | Sources describing a "staking" feature conflict — one points to a third-party platform, another to an uncorroborated validator model, and another to agent compute payments — so no reliable native mechanism description exists. |
| Islamic Contract Classification | 20/100 | The clearest staking-like description involves collateralized borrowing against a "fixed-income asset" on a third-party platform, which raises Qard-with-increment-style concerns, but this cannot be confirmed as a base-protocol feature. |
| Rewards Structure | 20/100 (low evidence) | No consistent, reliable source establishes the actual source or structure of staking rewards. |
| Documentation | 15/100 | Freysa's own documentation pages describe the agent framework and API but contain no dedicated staking terms, risk disclosures, or lock-up documentation. |
| Shariah Alignment | 18/100 | The contradictory and undocumented nature of any staking mechanism leaves a high degree of unresolved uncertainty (gharar) that cannot be assessed positively. |
Summary: Sources give conflicting and unverified accounts of any staking feature, with no official documentation establishing its terms, custody, or reward source.
Overall Assessment: Freysa/FAI is a real, funded, and actively developed AI-agent project, but its core value-generating mechanism resembles a chance-based prize-pool game, its distribution and staking claims are internally inconsistent, and no independent audit could be confirmed.
Written by ThanvirFounder, Ex Director S&P Global Energy
Reviewed by Imam Omar SiddiqiShariah Scholar