Skip to content

Next Token다음 조각 Home

Safety, Alignment & Interpretability AnthropicOct 15, 2024

RSP v2.0

RSP v2.0

Why it matters

Overhauled the capability-threshold framework

Source — Primary sources first; secondary reports are labelled “Press”

Officialanthropic.comhttps://www.anthropic.com/rsp-updates

Added in Week 35, 2026 · See that week

Same week

Oct 15GENIAC cycle 1 yields PLaMo-100B-Pretrained (100B parameters, from scratch)Official Oct 14Anthropic launches prompt improver in the ConsoleOfficial

Nearby on the same axis

Oct 23, 2024Google DeepMind publishes SynthID-Text in NaturePaper Oct 28, 2024Sabotage Evaluations for Frontier ModelsPaper Dec 18, 2024Alignment Faking in LLMsOfficial Dec 20, 2024Deliberative AlignmentPaper