ENTITY PROFILE / KNOWLEDGE GRAPH
Claude Opus 4.7
Eko is a copy of Claude Opus 4.7 Anthropic calls the concept 'Conway's Law' for agents to describe agent architecture. Claude Opus 4.7 scored 0% on ProgramBench. UK AI Security Institute tested Claude Opus 4.7 for cheating behavior in July 2026.· Entity summary generated by Graphiti
RELATIONS
4RelationsCURRENT
4CurrentEVENTS
4EventsREPORTS
4Source reportsNo public neighbors yet
Explore all published links of a node; previously discovered relations remain visible.
Relation timeline
BY VALID-FROM DATE- Claude Opus 4.7SCOREDProgramBench
Claude Opus 4.7 scored 0% on ProgramBench.
- AnthropicCALLSClaude Opus 4.7
Anthropic calls the concept 'Conway's Law' for agents to describe agent architecture.
- EkoIS COPY OFClaude Opus 4.7
Eko is a copy of Claude Opus 4.7
- UK AI Security InstituteTESTEDClaude Opus 4.7
UK AI Security Institute tested Claude Opus 4.7 for cheating behavior in July 2026.
Relations are facts Graphiti extracted from reports, not verbatim quotes. An end date means later reporting showed the relation no longer holds.
Events 4
- Why AI Models Cheat: Reward Hacking, Explained1 cited report · 2026-10-31
- Import AI 478: Math for alignment; ProgramBench; and Mirendil shows what RSI means1 cited report · 2026-10-26
- Anthropic says one AI agent beats many subagents now that models can handle complexity alone1 cited report · 2026-10-25
- AI researchers stranded Claude on a virtual desert island. It started to “play.”1 cited report · 2026-10-19
Source reports 4
- Why AI Models Cheat: Reward Hacking, ExplainedDecrypt · 2026-10-31
- Import AI 478: Math for alignment; ProgramBench; and Mirendil shows what RSI meansImport AI · 2026-10-26
- Anthropic says one AI agent beats many subagents now that models can handle complexity aloneThe Decoder · 2026-10-25
- AI researchers stranded Claude on a virtual desert island. It started to “play.”MIT Technology Review · AI · 2026-10-19