한국어EN
Safety, Alignment & InterpretabilityAnthropic, OpenAIAnthropicOpenAI

Anthropic and OpenAI publish a mutual alignment evaluation pilot

Mutual alignment evaluation pilot published

Why it matters

The first cross-company safety evaluation between competitors

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

Officialalignment.anthropic.comhttps://alignment.anthropic.com/2025/openai-findings

Added in Week 35, 2026 · See that week

Same week

xAI releases grok-code-fast-1OfficialOn the Theoretical Limitations of Embedding-Based Retrieval publishedPaperGemini 2.5 Flash Image (nano-banana)OfficialUN General Assembly adopts resolution A/RES/79/325 by consensusGovernment

Nearby on the same axis

xAI publishes its Risk Management FrameworkOfficialOpenAI and Apollo Research stress-test anti-scheming trainingPaperFrontier Safety Framework v3.0OfficialPetriOfficial