Crypto companies apply for new security tool from Anthropic

Some crypto companies, including Ethereum client developer Nethermind and wallet ZEUS, have applied to use the new OSS Scanner security tool from Anthropic. This service, launched on Thursday, is designed to provide vulnerability reports for open-source projects using the most powerful AI models, such as Claude Mythos. OSS Scanner will enable developers to find and fix vulnerabilities in code before they can be exploited by malicious actors. Anthropic stated that the reports will be generated by its most powerful models to give open-source projects maximum defensive advantage. Each case will be evaluated individually, considering the importance of the project for infrastructure and user security.

Crypto companies face growing threats from malicious actors, and this year has seen incidents of attacks using artificial intelligence technologies. Bitcoin exchange provider Boltz suspended operations in August, stating that attackers are developing exploits faster than the team can close them. Crypto payment service PayPerQ reported repeated attacks, suspected to involve AI. Anthropic warns that in the near term, AI may create an advantage for attackers: the cost of exploiting vulnerabilities is decreasing, while verification and fixing remain slow and dependent on humans.

Among those who applied for the OSS Scanner is VirtEngine, a decentralized cloud computing marketplace based on the Cosmos SDK. The program also includes developers of AI assistants, security agent tools, and machine learning infrastructure. At the time of publication, none of the applications had been processed and approved.