Decentralised News · 2027 planning series · Reviewed 21 November 2026
What Matters
A game can issue fewer tokens while running out of operating cash. It can move tokens into a treasury without destroying any supply. It can distribute rewards successfully without proving that players can convert them into income. Those are different economic events, and an AI assistant should keep them in different ledgers.
Before automating an emission rate or a sink, define the objective, the authority to change it and the evidence that would show the change occurred. DN’s game-economy record separates total supply, treasury token inventory and usable cash. It models a declared scenario without predicting token prices, player earnings or the success of a game.
The Signal
Published game policies show why a permanent reward assumption is fragile. Axie’s January 2024 SLP announcement described a supply target and a stability-fund mechanism; it expressly said the cap was enforced through a social commitment and game emission mechanisms. Sky Mavis’s later history records that Origins SLP emissions stopped on 7 January 2026 because the reward structure enabled automated farming, while deposits for crafting and morphing remained available. These are dated policy decisions with a named scope, not evidence of an autonomous AI central bank. [1] [2]
DN Alpha Thesis
Automating a game economy is useful only when the system can explain which balance it changes and who approved the change. Supply, player claims, treasury inventory, liquidity and operating cash should have separate measures. A policy that improves one measure may worsen another. The decision needs a stated trade-off, not a single health score.
This article introduces an original DN planning method for game operators, researchers and people assessing token-economy claims. Sources were reviewed on 21 November 2026. No live game integration, bot-detection benchmark, player dataset or economic intervention was tested. The calculations are synthetic and do not establish permission to automate gameplay.
1. Start with an event ledger, not a token-price target
List the events that the game creates: a reward is awarded, becomes claimable, is transferred, is spent on crafting, is sent to a treasury, is destroyed under the contract’s supply rules or is exchanged for another asset. Give each event an identifier and a specific meaning. Moving a balance from pending to claimable should not automatically be counted as a second reward.
For an operator, a useful reconciliation links the game record to the token record and then to any relevant cash receipt. For a player, it distinguishes an estimated reward from an eligible claim and an actual receipt. For a researcher, it makes clear whether the reported metric measures events, accounts, tokens or money.
The ERC-20 standard distinguishes total supply, account balances and transfers. OpenZeppelin’s supply guide also explains that the standard does not prescribe how a token’s supply is created. Consequently, a familiar interface is not enough to infer the game’s emission policy or identify every supply-changing event. Inspect the deployed mechanism and the relevant period. [3] [4]
DN’s proposed event record includes game, mode, reward program, token contract, network, event time, event type, quantity and evidence reference. Use aggregate or nonpersonal records where possible. An economic reconciliation does not require collecting unnecessary player identities or biometric data.
2. Keep three ledgers with different units
| Ledger | Opening balance | Relevant changes | Question it answers |
|---|---|---|---|
| Total token supply | Declared total supply for the selected token and scope. | Newly minted tokens minus actual supply burns. | Did the token quantity increase or decrease? |
| Treasury token inventory | Tokens controlled by the specified treasury. | Token receipts minus transfers out and treasury-funded burns. | How many tokens remain available to that treasury? |
| Usable operating cash | Cash available for the stated obligations. | Actual or explicitly assumed cash receipts minus cash outflows. | Can the operator meet the cash obligations in the scenario? |
A transfer between two holders changes their balances without necessarily changing total supply. A treasury purchase can increase treasury inventory while reducing cash. A burn from the treasury can reduce both its token inventory and global supply. The same event may therefore appear in two ledgers for different reasons; it should not be double-counted within either ledger.
Define the scope before reconciling. A token bridged between networks may have representations that must be linked rather than added blindly. A rebasing token or a mechanism with unusual supply accounting needs a different model. This worksheet is intentionally limited to a simple, single-token mint/burn identity and does not silently generalize to every game asset.
3. A sink must specify where the tokens go
“Players spend tokens” is an incomplete description. The spend might transfer tokens to a developer, lock them temporarily, move them to a treasury, exchange them for an item or destroy supply. Those outcomes have different implications for inventory, future distribution and the quantity available to the rest of the system.
OpenZeppelin’s ERC-20 implementation documents a burn mechanism that reduces supply, as well as distinct capped and pausable extensions. That provides a technical vocabulary, not evidence that a particular game deployed those exact controls. A claim that a game has a hard cap should identify the contract and authority that enforce it. [5]
For a crafting sink, record the required token quantity, destination, supply effect and the item produced. Then ask whether the resulting item can itself generate rewards or be redeemed. The visible payment may be only one side of a larger economic obligation. A model that counts the payment as pure demand while ignoring the future reward commitment is incomplete.
The historical SLP announcement is instructive because it described a burn-linked allocation to a stability fund and made its implementation and policy discretion visible. Its terminology should not be copied into a generic formula without reading the described mint and burn sequence. The article uses that history as an example of why the ledger matters, not as a current SLP supply audit. [1]
4. Reward distribution is not self-funding
Mavis’s Reward Distribution documentation describes a service for ERC-20 payouts on Ronin, including distribution limits and a pending period. It also tells operators to maintain sufficient funds in the designated wallet. A distribution service can automate payment mechanics while leaving the operator responsible for the economic budget behind them. [6]
Keep the token budget separate from the cash budget. A game might have enough of its own reward token to pay a season while lacking cash for infrastructure, staff or contracted services. Conversely, a cash-funded operation could choose to limit token issuance for reasons unrelated to an immediate cash shortage.
Do not multiply a treasury’s entire token balance by a displayed price and label the result cash runway. A sale would need a permitted route, executable demand, costs and timing. The worksheet therefore has no token-price field. Any completed conversion enters the cash ledger as a separate receipt, with evidence and without also being counted in the opening cash balance.
The same distinction protects the reader evaluating an NPC earning claim. An entity receiving tokens has a token receipt. Whether it controls the account, may perform the task, can claim the reward and can receive usable money are further questions. DN’s existing gameplay-permission record addresses that separate authority boundary.
5. A changing policy needs an effective date and a version
Give each policy a name, version, scope and effective time. “Rewards were reduced” is too vague if one mode changes today while another retains an older schedule. Preserve the policy that applied to already earned or unclaimed amounts, and distinguish it from the policy for future activity.
The Sky Mavis history is useful precisely because it identifies Origins and a specific SLP change. It also describes later reward transitions in other named modes. A researcher should retain those distinctions rather than generalize one change to every Axie game or every reward asset. The documentation does not establish the economics of every subsequent season. [2]
DN proposes three separate states: proposed, approved and executed. A simulation can compare proposed settings. Approval records the authorized change. Execution requires evidence from the system that actually controls the policy. A successful presentation or governance vote should not be substituted for that last record.
Specify who may change a setting, the maximum change, any waiting period and the condition for human review. Those are operating choices for the actual project. A prompt telling an agent to preserve the economy is not an enforceable authority limit.
6. Measure the intended effect and the cost to players
A policy needs a measurable objective. Examples include reducing net supply growth, keeping the funded reward wallet above a threshold, lowering the number of unreconciled reward claims or preserving a chosen cash reserve. Each objective has a unit and an observation window. None by itself establishes that the game is enjoyable or commercially durable.
Record possible adverse effects before changing a rule. A more expensive crafting action could reduce token use if players stop crafting. A lower reward may change participation. A claim delay might reduce immediate outflows while increasing the number of outstanding obligations. These are hypotheses to evaluate, not effects that this article has measured.
Use comparison groups or staged changes where the project can do so appropriately, and document simultaneous changes that could explain the result. An increase in active wallets does not by itself establish an increase in unique people or paying customers. A larger volume number does not establish higher retained cash.
For automation-related enforcement, keep the model output separate from the decision and appeal process. A suspected bot classification does not establish fraud. A policy should not respond to an uncertain label by automatically imposing a financial penalty that the rules and evidence do not support. This article supplies no detection accuracy claim or biometric requirement.
7. AI should prepare bounded options and preserve the countercase
An assistant can organize event data, identify a supply reconciliation gap and compare scenarios. The useful prompt is specific: explain why these mints, burns and balances do not reconcile, or show what happens to cash if the entered receipt assumption is lower. The output should retain the records and assumptions used.
A proposed adjustment should include the objective, the changed parameter, the expected accounting effect, the uncertainty and the evidence that would trigger reconsideration. Preserve a countercase, such as lower participation or no token-sale cash receipt. That makes the decision reviewable without claiming the model can stabilize a market price.
Keep the ability to recommend separate from the ability to change production rules or spend treasury funds. A deployment also needs a human contact for exceptions and a way to stop new instructions. Stopping the agent does not undo rewards already promised or transfers already completed; those obligations need their own reconciliation.
Worked example: deflation and cash runway answer different questions
A fictional game begins a month with 10,000,000 tokens in total supply. During the synthetic month, 500,000 tokens are minted and 300,000 are actually burned. Closing supply is 10,200,000, an increase of 200,000 or 2%. Transfers into the treasury do not enter that formula.
The same treasury begins with 1,000,000 tokens, receives 100,000, transfers 200,000 out and burns 50,000 of its own tokens. Those 50,000 are already included in the global 300,000 burn figure. Its closing token inventory is 850,000. The model rejects a treasury burn larger than the total reported burn, because the declared scopes would not reconcile.
Operating cash begins at 30,000 currency units. Assume recurring receipts of 4,000 and recurring cash outflows of 12,000 per month. The cash deficit is 8,000 per month. Under those unchanged assumptions, cash covers 3.75 months, and a six-month horizon has an 18,000 funding shortfall. This is a constant-flow sensitivity, not a forecast of the game’s future.
Now increase the global burn assumption to 600,000 while keeping cash flows unchanged. Total supply falls to 9,900,000, but the cash runway remains 3.75 months. The counterexample shows why a deflation claim cannot answer whether the operator can pay its bills.
Alternatively, move another 200,000 tokens from players into the treasury without burning them. Treasury inventory rises while global supply is unchanged. The worksheet will not convert those tokens into cash. A separate completed sale or other cash receipt would need to be entered in the correct ledger.
Use the DN Game Economy Supply and Cash Record
Choose one period and one token scope. Enter opening supply, mints and actual burns for that period. Then enter the treasury’s own token receipts, transfers and burns; treasury burns must already be part of the global burn total. Finally, enter usable cash and the recurring cash-flow assumptions for the horizon.
The instrument separates a proposed policy from a user-declared approved or executed change. Selecting executed does not authenticate execution. Keep a nonpersonal evidence alias for the record and use the actual governance or operator process to establish authority.
DN Game Economy Supply and Cash Record
Change the illustrative assumptions below. This worksheet runs locally in your browser. It records your declarations; it does not verify evidence, connect to accounts, submit instructions or grant approval.
The initial numbers are synthetic. Unknown evidence remains unresolved. Downloads contain the values you enter; use aliases and avoid private or identifying information.
A zero or negative net cash outflow is shown as no depletion under the entered constant-flow assumptions. It is not described as permanent sustainability. Missing numbers are not treated as zero, and impossible negative supply or inventory states are rejected instead of being hidden by a positive-looking score.
Methodology and practical limits
The supply identity is opening supply plus mints minus burns. The treasury identity is opening inventory plus receipts minus transfers out minus its own burns. Cash depletion uses recurring outflows less recurring receipts; the runway divides usable opening cash by a positive deficit. The horizon shortfall is the greater of zero and cumulative assumed cash deficit less opening cash.
These equations exclude price formation, market depth, player response, liabilities not entered, legal treatment, exchange conversion, financing and intraperiod payment order. Monthly net cash can conceal a bill due before a receipt. Use an actual dated cash calendar before relying on a planning result for operations.
Refresh the record after a rule change, token upgrade, migration, reward-asset change or material difference between reported and observed balances. The outcome should be a reconciled event trail and a clearly owned next decision. An AI-generated economic narrative is useful only when a reviewer can trace it back to those records.
Related DN research
Check gameplay authority with the game automation permission record. Use the DAO treasury control framework for grants and spending authority, and the token-holder payout analysis before assuming that game revenue belongs to a token holder.
Sources and evidence register
Primary sources reviewed on 21 November 2026. The article distinguishes documented facts, DN analysis and synthetic examples. Recheck changeable rules and configurations before acting.
- Axie Infinity: SLP monetary policy, 11 January 2024 — Historical social/emission cap and burn-linked stability-fund design. Not a current supply audit or proof of autonomous AI management.
- Sky Mavis: Lunacian Homecoming history, updated 1 June 2026 — Records Origins SLP emissions cessation from 7 January 2026 and later named reward changes. Scope and effective dates must remain distinct.
- ERC-20 Token Standard — Separate total-supply, balance and transfer interfaces; no universal reward policy.
- OpenZeppelin: Creating ERC-20 Supply — Supply mechanisms are implementation choices; the interface alone does not determine policy.
- OpenZeppelin ERC-20 API documentation — Documented burn, capped and pausable mechanisms; no claim that any named game uses a particular implementation.
- Mavis: Reward Distribution — Documented payout mechanics, limits, pending period and wallet funding requirement; not a source of economic revenue.
Frequently asked questions
Does spending a token reduce its total supply?
Not necessarily. Spending may transfer tokens, lock them or invoke a supply-changing burn. Identify the actual mechanism and reconcile the event before classifying it.
Can a game be deflationary and still run out of cash?
Yes. Token supply and cash obligations are different ledgers. The synthetic example changes the burn assumption without changing the cash runway.
Does a treasury token balance count as operating cash?
The worksheet does not treat it that way. A conversion requires its own route, price, costs and receipt evidence. Do not count a token holding and its eventual converted receipt twice.
Does the article establish that AI can autonomously balance a game economy?
No. It provides an original planning and reconciliation method. There is no tested intervention, price forecast, player-response model or ranking of autonomous economies.
Does approval prove that a policy change executed?
No. Proposed, approved and executed are separate states. The worksheet records declarations and does not authenticate the underlying governance or production change.
Why does the tool reject inconsistent burn or inventory inputs?
A treasury burn must fit within the declared total burn, and a balance cannot become negative under the simple accounting model. An inconsistency requires scope or source reconciliation.