Anthropic Launches Built-In Memory for Claude AI Agents, Cutting Errors by 97% in Early Tests
Summary
Anthropic launches built-in memory for Claude AI agents in public beta, enabling cross-session learning that has already slashed first-pass errors by 97% and reduced latency by 34% for early adopters like Rakuten, while Wisedocs reports 30% faster document verification.
Key Points
- Anthropic is launching built-in memory for Claude Managed Agents in public beta, allowing agents to learn and retain information across sessions using a filesystem-based memory layer optimized for long-running performance.
- Memory stores are enterprise-ready, featuring scoped permissions, audit logs, version control, and full API access, enabling multiple agents to share and access memory concurrently without overwriting each other.
- Early adopters are already seeing major gains, with Rakuten reporting 97% fewer first-pass errors and 34% lower latency, and Wisedocs speeding up document verification by 30% using cross-session memory.