Anthropic is supercharging cyber teams with expanded access to its AI models, following a staggering discovery of 129,000 flaws - with over 33,000 rated critical or high-severity. The company's Cyber Verification Program now offers tiered access to advanced models like Claude Opus 5.5, empowering teams to tackle vulnerabilities head-on.
OpenAI’s GPT-6.2 Astra and Sol take first and third on DesignArena’s game dev leaderboard