Glossary
Slashing is a proof-of-stake network's penalty for provable validator misbehaviour: a slice of the staked collateral is burned. For anyone staking, it is the downside attached to the yield - honest software, honest uptime, and the penalties stay someone else's.
Slashing is the punishment a proof-of-stake network imposes on validators caught provably cheating. On Ethereum, a validator stakes 32 ETH as collateral, and a portion of that collateral is burned - permanently destroyed - when the validator signs two contradictory blocks or attestations. The burned coins go to an unusable address, removing them from supply. The validator is then force-exited from the network.
Slashing is not the same as downtime. A validator that simply goes offline is penalised with the inactivity leak, a slow drip that costs a fraction of a percent a month - annoying, not fatal. Slashing needs proof of contradictory signing, which is almost always a software misconfiguration or a duplicated key rather than an attack. Delegated stakers share the penalties of the validator they delegate to, which is why an operator's uptime record matters before the deposit goes in.
A validator's job is signing: proposing blocks and attesting to other validators' proposals. The signature is unique to one view of the chain, and signing two contradictory views - the same slot filled twice, or two conflicting attestations - is the crime slashing punishes. Any other validator can submit the evidence on-chain, and the protocol burns part of the offender's stake. The minimum penalty is one thirty-second of the stake, which is 1 ETH on a 32 ETH validator, roughly $3,790 AUD. The validator is then removed from the network over an exit queue.
The correlation penalty is the scaling part. The burn grows with how much of the network is slashed within a roughly 36 day window: one rogue validator costs the minimum, while a third of the network failing together would approach the full stake. The design intent is deterrence against coordinated attacks, because an attacker's own collateral burns with them. A slashed validator's coins are sent to an unusable address, permanently removing them from supply - deflation, at the offender's expense.
Delegation routes the risk through an operator. A pool combines many participants' stakes, runs the hardware and client software, and shares rewards and penalties pro-rata - so one operator's misconfiguration lands on every delegator at once. Downtime, by contrast, triggers the inactivity leak: an offline validator loses on the order of a fraction of a percent of its stake per month, which on 32 ETH is a few hundred dollars rather than thousands. The validator entry covers the honest-work side of the ledger.
For an Australian staker, slashing prices the risk inside the yield. A direct validator needs 32 ETH - roughly $121,300 AUD at recent prices - plus hardware, uptime, and correctly configured client software, and a botched migration is the classic way people meet the 1 ETH minimum. ASIC MoneySmart's guidance is blunt that staking carries loss risk on top of price risk, and slashing is the named mechanism.
A colleague named Brent (yes, that Brent) ran a second validator on the same keys as a backup and called it redundancy. It was the classic double-signing setup, and the only reason it never slashed is that he never powered both on at once. The highest-leverage things most Australians can do are simpler: stake through a vetted pool if running hardware is not realistic, check the operator's slashing history before depositing, and keep the ATO paperwork straight - rewards are ordinary income, and a slashed stake is a capital loss. The staking entry covers the income side, and the liquid-staking entry covers the receipt-token version.
A validator holding 32 ETH - about $121,300 AUD at ETH's recent $3,790 - is migrated between machines and briefly runs twice on the same keys. Both instances sign the same slot differently, the contradiction is submitted on-chain, and the protocol burns the minimum: 1 ETH, about $3,790, plus a correlation penalty while the exit queue processes. The validator leaves the network, and the remaining 31 ETH returns to its owner. The stake is not the only loss - the rewards stop the moment the slash lands, and neither disposal triggered the other.
The minimum burn is designed to grow with company. If 30 percent of Ethereum's roughly one million validators were slashed inside the same 36 day window, the correlation penalty approaches the full 32 ETH for everyone in that group - about $121,300 AUD each. No mass slashing has happened on Ethereum mainnet, and the handful of individual slashings on record were accidental double-signs. The arithmetic is the deterrent: an attacker funding an attack burns their own collateral with the crowd they recruited.
A validator loses internet for a month during normal network conditions and misses every attestation. The inactivity leak costs on the order of a fraction of a percent of the stake - roughly 0.05 to 0.1 ETH, a couple of hundred dollars - compared with slashing's 1 ETH floor. The validator wakes up, reattaches, and the drip stops. Downtime is expensive at the margins; slashing is expensive by design.
Mistake: Calling every penalty slashing.
Why it bites you: Downtime triggers the inactivity leak, a drip worth a few hundred dollars a month on 32 ETH. Slashing needs provable contradictory signing, which is a different and larger event.
Mistake: Running duplicate validator instances on the same keys.
Why it bites you: Two instances signing the same slot is the textbook slashable event. Brent's 'backup' validator was one power-on away from a 1 ETH lesson.
Mistake: Delegating to the highest-yield pool regardless of its record.
Why it bites you: Penalties flow through to delegators pro-rata. A slashed operator passes the burn to everyone who deposited, and the extra yield never priced that.
Mistake: Assuming liquid staking removes slashing risk.
Why it bites you: stETH and rETH holders share the underlying validators' penalties. The receipt token can also trade below its backing after an incident, as stETH did in June 2022.
I spent an afternoon reading the Ethereum slashing specification so you don't have to, and the detail that surprised me is the correlation penalty: the protocol deliberately makes misbehaviour more expensive in company. One validator's mistake costs the minimum; a coordinated failure approaches the full stake. It is a deterrent design, and a good one.
The recurring conversation I have with Brent is about his backup validator, which was the exact setup the specification warns about. He got lucky; the specification does not promise luck. My position is unchanged: run hardware only if you will read the client documentation, delegate to an operator with a clean record if you will not, and never call a slow month a slashing. The yield pays for the risk; the records pay for the tax office.
No. Offline validators pay the inactivity leak, which costs a fraction of a percent of the stake per month, while slashing requires provable contradictory signing. The two penalties are separate mechanisms in the protocol.
The minimum is one thirty-second of the stake - 1 ETH on a 32 ETH validator, roughly $3,790 AUD. The correlation penalty scales that up if many validators are slashed within the same window.
Yes - the underlying validators can be slashed, and the penalties flow through to delegators pro-rata. Liquid staking does not remove the risk; it changes who runs the hardware.
No. The individual slashings on record involve small numbers of validators, almost always from accidental double-signing or client misconfiguration rather than attacks.
The burned portion is a capital loss on a CGT asset, claimed with evidence, while rewards earned before the slash remain ordinary income in AUD. Keep the transaction records either way.
Disclaimer: This information is general in nature and does not constitute financial or legal advice. Always consult a qualified professional for your specific situation.

Financial Chaos Analyst
Ivy Sinclair-Wren is a Financial Chaos Analyst covering crypto markets, investing behaviour, and the Australian habit of checking prices during family dinners. Based in Melbourne, she reviews AUD Crypto Tracker's coin guides for accuracy, ATO alignment, and ruthless utility — no hype, no predictions, just what the data says.