Ethereum’s Phylax Security Tool Raises Censorship Concerns

Cointelegraph


Phylax’s new security system could help decentralized finance (DeFi) protocols stop malicious transactions before they settle.

That may bolster crypto’s defenses against North Korean hackers and United States-sanctioned actors, but it may also introduce a way to freeze smart contract functions without explicitly giving anyone the power to do so.

And that’s the kind of control that raises concerns the system could end up as a centralized tool for law enforcement rather than decentralized finance.

“The question is not just whether there is risk, but if it’s acceptable risk,” says Stefanie Boss, co-author of research on censorship resistance and compliance behavior in Ethereum’s consensus mechanism.

“In this case, I’d lean toward that it is acceptable, as the goal is to make the network safer and more aligned with the network and application values,” she says.

A security mechanism with a potential freeze risk

Announced on Nov. 10, Phylax’s integration with block builder Titan Builder, allows Ethereum applications to define rules that exclude transactions that may produce harmful outcomes during block construction.

Titan, which Phylax says produces one in every two Ethereum blocks, checks transactions against those rules before including them in its blocks.

The idea is to prevent exploits from settling rather than relying on monitoring systems or emergency interventions after an attack has begun. “Our mission: stop users from losing money to hacks and errors in DeFi,” the announcement said.

Related: Fears of AI-driven DeFi hack epidemic overstated for now — but not for long

In a blog Phylax uses the scenario of a vulnerable vault that could mistakenly release $1 million to a user whose shares are only worth $1,000. With the system in operation, a circuit breaker could cap the amount leaving the vault in a single transaction, preventing the funds from being drained.

Similar safeguards could also potentially limit losses in major hacks like the $6.3 billion Binance hack or the $380 million Bitget exploit, where malicious actors drained exchange wallets and were able to move large amounts of crypto through decentralized protocols.

Phylax says hacks should not settle. Source: Phylax Systems.

While preventing malicious transactions would be a win for the crypto space, digital assets lawyer Gabriel Shapiro warned that applications using the system could become effectively freezable if builders stopped providing the required attestations.

He explained that smart contracts can be configured to reject transactions unless they are included in blocks from builders that have performed the required security checks. If those builders stopped providing attestations, protected functions could keep rejecting transactions until they resumed.

Related: DeFi TVL drops 39% in 2026 amid market downturn and record hack activity

Phylax founder and chief executive Odysseas Lamtzidis says protocols can avoid that risk by allowing transactions to go through even when the system’s security checks are unavailable.

“The sane default we provide and suggest to protocols is a fail-open behavior,” he tells Magazine.

But that safeguard is optional, leaving individual protocols to decide how much priority to give security over availability. “Some protocols might choose to prioritize safety over liveness,” Lamtzidis says.

He also says protocols can decide whether their smart contracts are immutable or can be changed through governance.

Shapiro says fail-open protections should be mandatory and changes to the safeguards should be subject to time delays to prevent protocols from removing them immediately.

Ethereum’s builder concentration problem

There is also a broader concern about whether the system could reinforce Ethereum’s already heavily concentrated block-building market.

Fan Zhang, an Ethereum researcher, says Phylax’s technology is only becoming practical now because of that concentration. “We finally have a way to implement this effectively,” he says, “but sadly only because of how centralized Ethereum’s builder market has become.”

Zhang says Phylax’s approach has “appealing security benefits, especially in the age of AI bots being able to find bugs faster than ever.”

But he warns that introducing additional code at such a critical point in transaction processing creates new failure modes, including implementation errors and builder signing-key theft, loss or impersonation.

“I think the community is very concerned about centralized builders and is making efforts to address them. So I’m concerned that these efforts may undermine each other,” he adds.

Related: FBI blames North Korea for record $6.3B Binance hack

Ethereum researcher Sen Yang tells Magazine, “If major protocols rely on a small number of builders for security enforcement, this could give those builders a competitive advantage and further exacerbate market concentration.”

Phylax announces its circuit breaker system. Source: Phylax Systems

He says Ethereum’s proposed enshrined proposer-builder separation (ePBS) upgrade, slated for inclusion in the upcoming Glamsterdam upgrade, may reduce reliance on third-party relays, “but it does not directly address builder market concentration.’’

Yang says transactions excluded by dominant builders could still be included by non-censoring builders, potentially causing delays rather than permanent exclusion.

Where does security become censorship?

Phylax argues that its system does not introduce a new authority over Ethereum transactions because applications themselves define the rules. The technology is “explicitly opt-in,” Lamtzidis says.

He compares the checks to conditions already written into smart contracts. “It’s no different than a ‘require()’ statement,” he says. “Censorship implies arbitrary decision-making or coercion.”

Zhang makes a similar distinction between applications voluntarily filtering transactions and external parties imposing censorship. “Applications willingly signing up for filtering is fine, as a user can walk away from such applications,” he says. “It’s externally forced censorship that’s the problem. That’s where I’ll draw the line.”

Boss says rules targeting a transaction’s outcome are less concerning than those targeting who is making it. “In this case, it is also result-driven, which is often considered less risky,” she says.

She considers the security objective legitimate, arguing that “the harm of those hacks is much higher than the harm of this type of exclusion.”

But she warns that rules could also be used to enforce sanctions or politically motivated exclusions.

“It is important to ensure that there are proper safeguards in place to make sure that these rules will remain fair and designed for security,” she says.

Magazine: Banks are banning AI agents over security… and competition concerns



Source link