The vulnerabilities could allow malicious peers to overwhelm Bitcoin nodes, while some older systems remain exposed despite the latest patches.
Bitcoin Core disclosed three AI-discovered security vulnerabilities that could disrupt nodes, including one allowing attackers to remotely crash Windows and Mac systems.
The Oct. 28 disclosures identified two medium-severity memory-exhaustion vulnerabilities and a low-severity file-overflow flaw, exposing weaknesses in software used to validate transactions and maintain the Bitcoin network.
Developers released fixes in versions 29.5, 30.4, 31.2 and 32.1 after automated large language model (LLM) scanning discovered them. The findings highlight artificial intelligence's growing ability to identify security weaknesses in critical cryptocurrency infrastructure.
The most immediately exploitable flaw allows a malicious peer to overwhelm Bitcoin nodes running on Windows, macOS, and BSD by sending specially crafted network data.
According to Bitcoin Core's advisory, the vulnerability bypasses safeguards designed to limit incoming data, allowing memory consumption to increase until the affected node crashes.
Linux systems are unaffected by this particular flaw because they use a different mechanism for monitoring network connections.
A second medium-severity vulnerability affects how Bitcoin Core communicates with older peers that do not support a network feature introduced under Bitcoin Improvement Proposal 31.
Thanks for signing up to The Catalyst. Look out for your first daily email. If you already receive it, you’re all set.
Malicious peers could exploit the mechanism to accumulate outgoing messages indefinitely, increasing memory consumption by up to 2 megabytes per hour per inbound connection and 5 megabytes per hour per outbound connection.
Source link
