Buterin argues Minimmit offers better security tradeoffs than Casper FFG for Ethereum finality. Here’s what he said and why it matters.
Vitalik Buterin has weighed in on a key shift in Ethereum’s consensus design.
In a post shared on X, he outlined why Minimmit could replace Casper FFG as Ethereum’s finality gadget. The proposal involves trading a small drop in fault tolerance for stronger protections against the worst-case attacks.
Buterin, who has long pushed for high security standards in Ethereum consensus, says this tradeoff is not just acceptable but preferable.
Minimmit vs Casper FFG: How the Two Systems Compare
Casper FFG uses two voting rounds to finalize blocks.
Validators sign once to “justify” a block and again to “finalize” it. Minimmit cuts this down to a single round. In exchange, its fault tolerance drops from 33% to 17%.
This means an attacker needs to control just 17% of staked ETH to halt finality under Minimmit, compared to 33% under Casper FFG. At first glance, that sounds like a step backward. Buterin, however, argues the full picture tells a different story.
He points out that stopping finality is not the most dangerous outcome on a blockchain. Censorship attacks, he says, are far worse.
Both systems require an attacker to hold 50% of staked ETH to pull off censorship. On that front, neither protocol gives ground.
Why Vitalik Thinks Censorship Is the Bigger Threat
Buterin draws a sharp line between finality reversion and censorship.
Reverting finality leaves behind massive on-chain evidence. That evidence can trigger slashing, costing the attacker billions of dollars. Censorship, on the other hand, requires social coordination to counter and leaves no automatic punishment trail.
Under Casper FFG, an attacker holding more than 67% of staked ETH can unilaterally finalize blocks while censoring.
Under Minimmit, that threshold rises to 83%. This is a meaningful difference. The higher the bar, the harder it is for a censoring attacker to also lock in their preferred chain.
Buterin also raises the scenario of a software bug. If a critical bug reaches 67% of clients under Casper FFG, bad data could finalize.
Under Minimmit, that same bug would need to spread to 83% of clients before anything finalizes. More room for error means more time to catch and fix problems before they become permanent.
Related Reading: Vitalik Buterin Urges Ethereum to Rethink Apps, AI, and Privacy
The Case for Two Chains Over Wrong Finality
Buterin frames his argument around a core principle: finality should mean final.
When things go wrong, the better outcome is two chains running in parallel rather than one wrong chain locking in. Minimmit expands the range of situations where that “two chains dueling” outcome occurs, giving validators and developers time to identify what went wrong.
He also addresses the economic argument directly.
With around 15 million ETH staked, reverting finality under Minimmit would still require slashing roughly 17% of that, or close to 2.5 million ETH. At current prices, that translates to billions of dollars in losses for an attacker.
Buterin argues that if someone has that kind of capital and wants to cause damage, there are far more effective ways to use it than attacking Ethereum.
Discussing fast finality with people, it is the move to 1 voting round that causes most concern since it ostensibly reduces security from 33% to ~17%.
Here V explains why this is fine, and maybe even better than the status quo.
DMs remain open if you want to talk fast finality… https://t.co/rSPWMaFTcr
— Ben Edgington (@benjaminion_xyz) March 6, 2026
Ben Edgington of the Ethereum consensus team echoed this framing on X, noting that the move to a single voting round causes the most concern in community discussions.
He called on bridges and interoperability platforms to reach out and discuss fast finality, signaling that the conversation is still open for input across the ecosystem.
