What is Aave (AAVE)?

Aave is a lending market with no loan officers. Deposit an asset and it joins a pool that borrowers draw from; borrow against collateral worth more than you take. Interest rates are not negotiated — they move with utilisation, rising as a pool empties to attract deposits and pull borrowing back. Everything runs in contracts, and positions that fall below their collateral threshold are liquidated automatically by anyone willing to do it for the incentive.

Over-collateralisation is what makes lending possible without identity or courts. You cannot borrow more than you post, so there is no credit assessment and no recovery process — just a liquidation threshold and a bot watching it. It also means Aave cannot do the most useful thing banks do, which is lend to people who do not already have the money.

What Aave pioneered

Flash loans came from here: borrow any amount with no collateral at all, provided you repay within the same transaction. If you do not, the whole transaction reverts as though it never happened, so the protocol cannot lose. It sounds like a loophole and it is a genuine primitive — used for arbitrage, collateral swaps and refinancing, and also the tool of choice in a long list of exploits against other protocols.

GHO, and why it matters to the token

GHO is Aave's own over-collateralised stablecoin, minted against collateral already deposited in the protocol. The significance is economic rather than technical: interest paid on GHO goes entirely to the DAO rather than to depositors, which makes it a much higher-margin product than lending. Under the Aave Will Win framework, 100% of revenue from the protocol, GHO and Aave-branded products flows to the DAO treasury.

Aave v4

The current architecture consolidates liquidity under a unified design rather than fragmenting it across separate markets, and reached around $900m during its 2026 migration phase. Consolidation is the theme: fewer, deeper pools rather than a proliferation of isolated ones.

• Governance over listings, collateral parameters, risk settings and the treasury.
• Staking in the Safety Module, which backstops the protocol against a shortfall event.
• The buyback, which routes all protocol and GHO revenue into open-market AAVE purchases automatically.
• Borrowing discounts on GHO for stakers.

Why Aavenomics 3.0 is different from other buybacks

Most token buybacks are a committee's decision, renewable and revocable — Optimism's, for instance, is a twelve-month pilot governance can decline. Aave's, since 27 June 2026, is immutable and non-discretionary: revenue routes to purchases without anyone signing off, and the mechanism cannot simply be switched off by a vote of the people it benefits.

That is a meaningful design difference. It removes the discount a market applies to a promise someone can withdraw. It also removes flexibility — if the protocol needs that revenue for something else in a crisis, the mechanism does not care. Aave chose credibility over optionality, deliberately, and the results are measurable: 292 AAVE a day, over 205,000 acquired, more than 1.28% of supply.

16 million AAVE is the maximum supply, reached through the 2020 migration from LEND at 100:1. A portion sits in the ecosystem reserve, governed by the DAO. There is no ongoing issuance beyond what governance votes for.

Against that fixed ceiling, the buyback is a steady one-directional flow. At roughly 292 AAVE a day the annual rate is around 106,000 tokens, or about 0.7% of maximum supply, funded from roughly $402m of annualised protocol fees. Aavenomics 3.0 also cut DAO spending, which matters because a buyback funded alongside rising expenses is a different proposition from one funded alongside falling ones.

What the revenue actually is

Aave earns a share of the interest borrowers pay, plus liquidation fees, plus the full interest on GHO. The GHO line is the highest-margin and the fastest-growing, and it is the reason Aave's economics improved faster than its lending volume did. Under Aave Will Win, all of it reaches the DAO rather than being split with anyone.

Staking AAVE means depositing into the Safety Module, which is insurance rather than yield. If the protocol suffers a shortfall — bad debt from a liquidation that fails, or an exploit — up to 30% of the staked pool can be seized to cover it. In exchange, stakers earn rewards and a discount on GHO borrowing.

This is a genuinely different risk from proof-of-stake staking and it gets conflated constantly. On Ethereum your stake is at risk from your validator's misbehaviour. Here it is at risk from the protocol's losses, which you do not control and cannot monitor. You are underwriting Aave's credit book.

There is also a cooldown before unstaking, followed by a window to withdraw. Miss the window and you start again — which matters, because the moment you would most want to exit is the moment the Safety Module is most likely to be needed.

Bad debt is the structural risk

Over-collateralisation works until a price moves faster than liquidators can act, or until an asset's on-chain liquidity is too thin to sell the collateral into. Aave has taken bad debt before, most notably from a manipulated illiquid collateral market in 2022. The Safety Module exists precisely because this can happen and will again.

Oracle dependence

Every liquidation depends on a price feed. A wrong price liquidates solvent users or fails to liquidate insolvent ones, and most of the worst DeFi failures have been oracle failures rather than contract bugs. Aave relies heavily on Chainlink, which is well-run infrastructure and a concentrated dependency.

The buyback cannot be turned off, which cuts both ways

Immutability is why the market trusts the mechanism. It also means that in a period where the DAO needs capital — to recapitalise after a shortfall, to fund a competitive response — the revenue keeps going to purchases regardless. Credibility bought with flexibility is still a trade.

Revenue is cyclical

Roughly $402m of annualised fees reflects a market with healthy borrowing demand. Lending revenue collapses in a bear market, because borrowing demand is leverage demand. The buyback scales down exactly when the token price is weakest, which is the opposite of when support is wanted.

Competition and margin

Morpho and others compete by returning more of the spread to users, and lending is not a business with much product differentiation. Aave's advantages are depth, integrations and a decade of risk parameters that have been tested. Those are real and they are not a moat against a protocol willing to run thinner.

Governance sets the risk parameters

Which assets can be collateral, at what ratios, with what caps, is decided by vote. A bad listing is how lending protocols lose money. Aave's risk process is among the most professional in DeFi, and it is still a DAO vote at the end.