Reading up on SWE-bench
6 deep · digging since nov 20, 25
- Introducing FrontierCode
Cognition's FrontierCode benchmark measures code mergeability, finding even top models like Claude Opus 4.8 score only 13.4% on its hardest 50 tasks.
- Anthropic says 80% of its new production code is now authored by Claude — how your enterprise can keep up
Anthropic reports that over 80% of its production code is now written by Claude, offering enterprises a three-step plan to adopt AI-driven development with cultural and verification overhauls.
- When AI builds itself \ Anthropic
Anthropic presents internal data showing AI systems now write 80% of its code and accelerate research, trending toward autonomous recursive self-improvement.
- Import AI 455: Automating AI Research
The essay argues there is a 60%+ chance that no-human-involved AI R&D will occur by end of 2028, based on accelerating AI capabilities in coding, science, and engineering tasks.
- What Jobs Are Made Of - by Thomas Wolf - Thomwolf
AI benchmarks show rapid progress but economic impact remains limited because jobs require judgment and agency, not just task execution — explaining why entry-level hiring is falling while senior roles grow.
- Building more with GPT-5.1-Codex-Max
OpenAI releases GPT-5.1-Codex-Max, a faster coding model using compaction for long-horizon tasks across multiple context windows.