As Sirer recently posted on X, “We will see AI exploiting system-level bugs long before ECDSA goes away.”
In his view, more advanced AI models can find the bugs in the blockchain software, libraries, and crypto wallets that have not been seen by humans.
Attackers may use these bugs to steal billions, he warned, long before AI breaks the widely used cryptography.
“If you believe, for instance that Ripple’s (or put any other high-value, low-use chain here) under-scrutiny code has no 0-day bugs that are hard for humans to see but easy for AI to detect, you’re a deluded person,” he stated.
Here, he is referring to a vulnerability in software that no one knew about before and that can be exploited by attackers before developers patch it.
Sirer, however, did not point to any specific $XRP Ledger vulnerability that has not been patched or demonstrate how AI attacked the network.
His statement comes as security issues of $XRP Ledger are back in the spotlight as the developers of the protocol rolled out an emergency software update on Sept. 25. Initially, the developers of $XRP Ledger did not release the source code of the security fixes. However, they promised to do it along with a technical retrospective.
Earlier, Ripple CEO Brad Garlinghouse also took aim at Sirer after the latter took aim at Ripple’s alleged lack of traction with banks.
AI threat sparks heated crypto debate
As reported by U.Today, Ethereum co-founder Vitalik Buterin recently issued a warning that AI-driven mathematical discovery could attack assumptions in lattice-based cryptography in 2 years.
Cardano founder Charles Hoskinson had a different take on Buterin’s worries. He called them speculative and said that math could not back up such predictions.
Ethereum researcher Justin Drake expressed concern over AI’s potential to one day attack Elliptic Curve Digital Signature Algorithm (ECDSA), a widely used method of signing crypto transactions.
Sirer admits that this cannot be fully dismissed but thinks the crypto community should be concerned about software vulnerabilities that already exist instead of only focusing on the potential breakthroughs in cryptography.