Article: Privacy is Fungibility: Why Endogenous Tokens Are Not Money
Date: 2026-05-15
Publisher: arXiv - Alex Lynham, Geoffrey Goodell
Score: ₿₿₿
Read time: 12-15 min
Summary:
- Overview: The paper argues that most endogenous blockchain tokens should not be considered money. The authors build on prior research that defines money through privacy and fungibility rather than simply through exchangeability. They examine public, permissionless blockchain systems and conclude that account-based ledgers expose users to forms of surveillance and control that are inconsistent with cash-like money. The paper also argues that stablecoins do not automatically solve this problem because they often inherit the same ledger structure and visibility. A central claim is that privacy and fungibility are closely linked and that public observability weakens both. The authors conclude that many cryptoassets function more like forms of credit than money.
- Privacy Enables Fungibility: The paper treats privacy as a core requirement for fungible money. When transaction histories are visible and traceable, units of the same asset can be valued differently based on their past activity, reducing true fungibility.
- Endogenous Tokens As Credit: The authors argue that native blockchain tokens rely on network rules, validators, and ledger state for their existence and enforcement. Because balances are maintained through account-based systems rather than private bearer instruments, these assets resemble credit relationships more than cash.
- Stablecoins Do Not Escape: The paper challenges the assumption that stablecoins are inherently closer to money than native cryptoassets. If a stablecoin depends on the same transparent ledger and shared state as the endogenous token securing the network, many of the same privacy and fungibility limitations remain.
- Tokenomics Design Implications: For token designers, the paper suggests that utility, governance, and price stability alone are insufficient for monetary use cases. Long-term value capture may depend on whether a system can support cash-like properties such as privacy, fungibility, and resistance to selective exclusion. Projects seeking to become monetary assets may need to focus as much on transaction architecture and privacy guarantees as on supply schedules or incentive mechanisms.
Article: Optimal Control of the Ethena Yield-Bearing Stablecoin
Date: 2026-05-11
Publisher: arXiv - Matthew Lorig
Score: ₿₿₿
Read time: 12-15 min
Summary:
- Overview: This paper analyzes the core yield-generation strategy used by Ethena's stablecoin system. The strategy combines a long position in staked Ethereum (stETH) with an equal-sized short position in ETH perpetual futures to create a delta-neutral portfolio that earns yield. The author models the problem as a stochastic control framework and derives optimal portfolio management rules. The goal is to maximize expected returns while controlling risk and maintaining stablecoin backing. The paper provides a mathematical framework for understanding how yield-bearing stablecoins can manage exposure under changing market conditions. It focuses on the economic mechanics behind sustainable yield generation rather than speculative token appreciation.
- Ethena Yield Engine Mechanics: The study formalizes the strategy that generates yield through staking rewards while offsetting directional ETH price risk using perpetual futures. This highlights that Ethena's yield is tied to real market income streams rather than relying solely on token emissions.
- Risk And Control Framework: The paper treats portfolio allocation as an optimal control problem. The framework seeks to balance yield maximization against changing market risks, funding rates, and portfolio constraints.
- Implications For Stablecoin Design: The research suggests that sustainable yield-bearing stablecoins require active risk management rather than static collateral structures. Performance depends on how effectively the protocol adjusts positions as market conditions evolve.
- Tokenomics And Value Capture: For tokenomics, the paper is notable because it studies a model where yield comes from underlying economic activity instead of inflationary token rewards. The long-term attractiveness of such systems depends on staking yields, derivatives market conditions, and risk-adjusted portfolio management. This points toward value capture models that are tied to cash-flow generation rather than continuous token issuance.
Article: Who Restores the Peg? A Mean-Field Game Approach to Model Stablecoin Market Dynamics
Date: 2026-01-26
Publisher: arXiv - Hardhik Mohanty, Bhaskar Krishnamachari
Score: ₿₿₿
Read time: 18-22 min
Summary:
- Overview: This paper studies how fiat-backed stablecoins return to their target price after a de-peg event. The authors build a model in which arbitrageurs and retail traders interact across primary and secondary markets. The goal is to identify which participants help restore the peg and under what conditions they act. The model is calibrated using stablecoin market data and tested against historical de-peg events. Results suggest that peg recovery is driven by a combination of arbitrage incentives, market structure, and trader behavior. The paper provides a framework for understanding stablecoin stability as an outcome of strategic market interactions rather than automatic market correction.
- Arbitrageurs Drive Peg Recovery: The analysis finds that arbitrageurs play a central role in bringing stablecoins back toward their target value. Their willingness to mint, redeem, and trade depends on expected profits, transaction costs, and access to primary-market mechanisms.
- Market Structure Matters Greatly: Peg restoration is influenced by how efficiently traders can move between primary and secondary markets. Frictions such as redemption delays, liquidity constraints, and transaction costs can slow recovery and increase the duration of de-peg events.
- Retail Traders Affect Stability: Retail participants contribute to market dynamics even when they are not performing direct arbitrage. Their reactions to price deviations, perceived risk, and market sentiment can either support or delay the return to the peg.
- Tokenomics Design Implications: For stablecoin designers, the paper suggests that peg stability depends heavily on incentive alignment for arbitrage capital. Redemption access, liquidity depth, and fee structures may be more important than simple collateral ratios during stress periods. Protocols that reduce market frictions and maintain profitable arbitrage opportunities are likely to achieve faster and more reliable peg recovery.
Article: Operating-Layer Controls for Onchain Language-Model Agents Under Real Capital
Date: 2026-04-28
Publisher: arXiv - T.J. Barton, Chris Constantakis, Patti Hauseman, Annie Mous, Alaska Hoffman, Brian Bergeron, Hunter Goodreau
Score: ₿₿+
Read time: 8-10 min
Summary:
- Overview: This paper studies how AI agents can manage real cryptocurrency capital safely and reliably. The authors analyze a 21-day deployment where 3,505 user-funded agents traded real ETH onchain. The system processed about 7.5 million agent invocations, 300,000 onchain actions, and roughly $20 million in trading volume. The main finding is that reliability came from the operating layer around the language model rather than from the model alone. Important controls included structured user settings, policy validation, execution guards, memory systems, and monitoring tools. The paper argues that financial AI agents should be evaluated across the full path from user intent to final transaction settlement rather than only through model benchmarks.
- Operating Layer Drives Outcomes: The study finds that infrastructure and control systems were the main source of reliability. Prompt compilation, typed controls, validation rules, and execution safeguards reduced harmful behavior and improved transaction success rates.
- Observed Agent Failure Modes: Testing uncovered several recurring problems, including fabricated trading rules, excessive concern about transaction fees, numeric anchoring, cadence-based trading behavior, and incorrect interpretation of tokenomics. These issues were difficult to detect through standard text-only evaluations.
- Control Improvements Increased Reliability: After targeted system changes, fabricated sell-rule behavior fell from 57% to 3%. Fee-driven observations dropped from 32.5% to below 10%, while capital deployment increased from 42.9% to 78.0% within affected test groups.
- Implications For Crypto Agents: The paper is relevant to tokenized AI agents and autonomous treasury systems because it shows that governance and control frameworks are as important as model quality. For projects building onchain agents, value capture and risk management may depend more on operating-layer design than on choosing a stronger language model.
Article: RSDM: The Consensus Honest Money in the AI Era
Date: 2026-05-01
Publisher: arXiv - Boliang Lin, co-author not identified in available abstract
Score: ₿₿+
Read time: 10-12 min
Summary:
- Overview: This paper proposes a monetary framework called Redeemable Self-Decaying/Devaluing Money (RSDM). The authors argue that AI-driven economies and cross-border autonomous agents require a globally accepted store of value that is resistant to fiat currency depreciation. RSDM is designed as a tokenized commodity money backed by physical assets such as precious metals. The key idea is that the recorded claim on the underlying commodity gradually decreases over time, reflecting storage and custody costs. The paper presents five online and offline issuance models for RSDM. The authors position RSDM as a modern form of “honest money” intended for long-term value storage in an AI-enabled economy.
- Commodity Backing With Decay: RSDM is backed by physical commodities rather than relying solely on government issuance or algorithmic monetary policy. Instead of charging explicit storage fees, the asset claim itself gradually declines over time, embedding custody costs directly into the monetary unit.
- Solving Storage Cost Problem: The paper argues that commodity-backed currencies historically faced challenges because storing and safeguarding assets creates ongoing costs. RSDM attempts to account for these costs through programmed self-devaluation, creating a mechanism that aligns the token's value representation with real-world storage expenses.
- AI Economy Monetary Design: The authors believe AI agents operating across borders will increasingly require a common unit of account and store of value. They argue that currencies exposed to persistent inflation may be unsuitable for long-duration autonomous capital allocation, creating demand for alternative monetary frameworks.
- Tokenomics And Incentives: From a tokenomics perspective, RSDM introduces a negative-carry asset where holding costs are built into token design. This discourages passive hoarding while maintaining commodity backing and predictable monetary rules. If implemented onchain, value capture would depend less on emissions or governance incentives and more on trust in reserves, redemption mechanisms, and the credibility of the decay schedule.
