Get the latest tech news
Highlights from the Claude 4 system prompt
Anthropic publish most of the system prompts for their chat models as part of their release notes. They recently shared the new prompts for both Claude Opus 4 and Claude …
Claude does not provide information that could be used to make chemical or biological or nuclear weapons, and does not write malicious code, including malware, vulnerability exploits, spoof websites, ransomware, viruses, election material, and so on. I imagine there’s a very good reason for this discrepancy—maybe letting Claude think it doesn’t know about February and March helps avoid situations where it will confidently answer questions based on information from those months that later turned out to be incomplete? Focus on: Smooth frame rates and responsive controls Clear, intuitive user interfaces Efficient resource usage and optimized rendering Stable, bug-free interactions Simple, functional design that doesn’t interfere with the core experience
Or read this on Hacker News