Opt-in open-source projects will receive free periodic security scans with suggested fixes, the company said.
Anthropic is expanding its cyber defence programme with new tools to scan open source projects and a collaborative mission that brings its security expertise to big businesses.
The AI giant and its contemporaries have been creating advanced models with never-before-seen capabilities to identify and exploit vulnerabilities. These models have already hacked several government websites and businesses, while also arming bad actors with tools to significantly boost their ability to carry out large scale hacks or phishing attacks.
Earlier this year, Anthropic launched Project Glasswing, a first-of-its-kind joint mission, alongside its cyber-focused model Mythos 5, to collaborate with major companies and government organisations to protect themselves against these newer AI-enabled risks (the EU only received access to Mythos last month after weeks of discussions with the company).
According to Anthropic, under this project, the company and its industrial collaborators scanned hundreds of widely used open-source projects and privately reported their findings to involved organisations.
Now, it is expanding its efforts to launch an open-source software (OSS) scanner, an opt-in service similar to Google’s OSS-Fuzz.
Projects that are enrolled under this will receive free periodic AI-generated security scans from Anthropic’s “most capable models”, the company said. The reports will include a proof of concept of how the bug could be exploited, an explanation and a suggested fix where one is available.
The company is also launching a critical infrastructure defence programme which brings frontier Claude models and on-site engineers to select providers, including Accenture, CrowdStrike, Deloitte, PwC and Palo Alto Networks among the project’s other founding members.
Several partners are currently working with Claude to fix vulnerabilities and help customers do the same, the company said.
Last month, Anthropic tapped Accenture to independently evaluate its frontier AI developments after CEO Dario Amodei called for better transparency across the sector. The partnership will be led by Accenture’s specialist AI business Faculty, which will conduct independent assessments and model safeguards testing at Anthropic.
Don’t miss out on the knowledge you need to succeed. Sign up for the Daily Brief, Silicon Republic’s digest of need-to-know sci-tech news.