Anthropic Research
Official Anthropic research papers and model safety updates.
Read through RSSHubAppears in:KI
Latest from Anthropic Research
- Teaching AI to build scientific modelsvor 3 Tg.
- The components of agent memoryvor 3 Tg.
- AI for scientific discovery: How researchers use Claudevor 4 Tg.
- Investigating vault particles with Claudevor 5 Tg.
- Progress on provable inference for large language modelsvor 7 Tg.
- Early progress on accelerating drug development with Claudevor 13 Tg.
- How AI agents shape experts’ judgment callsvor 13 Tg.
- An alignment assessment of the UK AISI cybertesting incidentvor 17 Tg.
- How Claude designed protein binders against scorpion neurotoxinsvor 20 Tg.
- Working at the frontier shapes expectations for the futurevor 24 Tg.
- How employment is changing at Anthropicvor 24 Tg.
- The AI expectations gapvor 24 Tg.
- Designing more efficient quantum circuits with Claudevor 24 Tg.
- Redesigning source control at Anthropicvor 27 Tg.
- Investigating unintended model actions in our evaluations and internal use2026-10-08
- The missing map of the sky2026-10-07
- Claude-shaped science2026-09-30
- What work can robots do?2026-09-29
- What do you want from AI?2026-09-28
- Project Swap: What happens when agents trade for us?2026-09-23