Critical Vulnerability in Solana Validators: AI Created Exploit

A critical vulnerability with a danger rating of 9.8 out of 10 has been discovered in Solana network validators. The issue affects the Linux kernel and allows attackers to remotely gain full control over the server.

The situation is complicated by the fact that an update for Ubuntu — the main OS used by validators — has not yet been released, although the fix is already prepared. Node operators are advised to immediately strengthen the security of their servers.

Anthropic's experimental AI model Claude independently created an exploit for this vulnerability. In 53 minutes, it gained full control over a test Linux server.

The main threat lies in the potential takeover of validator servers and access to private keys. This could disrupt the stable operation of the entire Solana network, especially if the attack affects a significant portion of nodes.