Ethereum client developer Nethermind and Bitcoin and Lightning wallet ZEUS are among several crypto companies that have applied to join Anthropic’s new OSS Scanner program, which offers open-source projects security reports generated by the company’s most powerful AI models.
Anthropic launched the opt-in service on Thursday. The stated goal is to help open-source developers find and fix vulnerabilities before attackers exploit them. The projects submitted their enrollment requests on Friday.
In its announcement, Anthropic said: “These reports will be generated by our strongest models, including Claude Mythos, to give open-source projects the largest defensive advantage.”
Crypto teams are pushing for frontier AI access
The applications arrive as crypto firms seek access to frontier AI models to defend against increasingly capable attackers. Cybersecurity experts have warned that uneven access could leave defenders at a disadvantage if stronger alternatives become more widely available.
Anthropic said OSS Scanner builds on its work on Project Glasswing. The company said it already scans open-source software for vulnerabilities and sends reports after human review, but added that manual review is slow, “so we’re not always able to share vulnerabilities as quickly as we would like.”
Under the new program, participating projects will receive vulnerability reports as soon as their code has been scanned.
Nethermind, ZEUS and VirtEngine are among the applicants
Pull requests in the OSS Scanner GitHub repository show that Nethermind has requested audits of its entire repository. ZEUS, a self-custodial Bitcoin and Lightning wallet, asked for its app to be reviewed for weaknesses affecting payments, private keys and connections to Lightning Services.
Another applicant is VirtEngine, a decentralized cloud computing marketplace structured as a Cosmos SDK chain.
Other applicants include developers working on AI assistants, agent-security tools and machine-learning infrastructure, as well as projects that provide software development tools, cloud storage and energy-system controls.
None of the pull requests had been merged at the time of publication.
Anthropic says projects will be reviewed individually
Anthropic said applications will be assessed case by case. The company said it will consider each project’s importance to infrastructure and user security, its exposure to remote attacks, and how many users or other projects depend on it.
Crypto companies have already reported AI-assisted attacks this year. Bitcoin swap provider Boltz suspended operations in August, saying attackers were developing exploits faster than its team could patch them. Crypto-payment service PayPerQ also reported repeated attacks that it suspected were AI-powered.
Anthropic warned on Thursday that AI may favor attackers in the near term because exploitation is becoming cheaper, while verifying and fixing vulnerabilities remains slow and dependent on people.
Related: Vitalik Buterin backs crypto “bunker mode” as AI math capabilities advance rapidly.