Mozilla leveraged the Claude Mythos preview and advanced harnessing techniques to find and fix 423 Firefox security vulnerabilities in one month—a 20x increase over their average—marking a qualitative shift in AI security auditing from noise generation to high-value signal production.
LlamaIndex launches ParseBench, the first document OCR benchmark for AI agents, alongside new parsing tools and benchmark results, marking a shift towards quantifiable document intelligence.
Anthropic's 'Project Glasswing', leveraging its latest AI model Mythos Preview, has helped partners discover over ten thousand high or critical-severity vulnerabilities in one month, shifting the core bottleneck in software security from 'finding vulnerabilities' to 'verifying and patching them'.
Anthropic, alongside Blackstone and Goldman Sachs, has formed a new company to provide deep customization and integration of Claude for mid-sized enterprises, signaling a new 'heavy delivery' phase in AI commercialization.
Anthropic launches Claude Design, an AI-powered collaborative design platform driven by Claude Opus 4.7, enabling designers to explore widely and non-designers to easily create professional visual work.