Research Jan 13, 2026
Anthropic Publishes Breakthrough in AI Safety Research
New Constitutional AI 2.0 paper demonstrates methods for training more honest, harmless, and helpful AI systems.
Research
Academic papers, breakthroughs, foundational AI research, and new discoveries.