AI Briefing#memory
The latest AI and developer news about #memory, with the original source and a short summary.
The latest AI and developer news about #memory, with the original source and a short summary.
The analysis argues that improvements in LLM agent capabilities stem from externalization, such as memory and skills, rather than weight modifications.
Participants using AI showed a 21% higher chance of correct decisions but experienced a 15.3% decline in unassisted performance by the fourth week.
A user test revealed that Claude's 'Use account memory' toggle is enabled by default for new projects and was found enabled on an existing untouched project, allowing non-project chats to access project-specific memory files.
LangChain introduced a file-based 'wiki memory' pattern that compresses raw data for AI agents.
HBF is discussed as a potential alternative to address DRAM capacity issues in LLM inference.
Reports indicate Nvidia is testing a 192GB-class low-capacity configuration for Rubin Ultra to address memory shortages.
Zero-Mem proposes a structure for retrieving and managing agent memory without LLM calls.
ChatGPT Work is the blueprint for the general-purpose knowledge work agent that OpenAI is aiming for.
SimpleMem compresses long-term memory to boost LoCoMo performance while cutting inference tokens by up to 30x.
An AI agent's memory is not mere knowledge storage, but a process of transforming past information into something useful according to a specific purpose.
A memory system that solves the session-memory problem for AI coding agents has been released.
As AI demand soaks up memory supply, price hikes for low-cost smartphones are now in full swing.
The open-source memory layer memv has launched an MCP server, making it available for use in Claude Desktop, Code, and more.
Anthropic unveiled a major update at Code w/ Claude centered on Claude Code, Memory, and Managed Agents.
Anthropic added Memory and Dreaming to its managed agents API.
Anthropic added dreaming, outcomes, and multi-agent orchestration to Managed Agents.
Vestige, an open-source MCP server that implements a cognitive memory layer for Claude, has been updated to v2.1.0.
Anthropic introduced Memory for Claude Managed Agents in public beta.
Released an AI memory that achieved Recall@5 59% on the LoCoMo-10 benchmark.
SK hynix has begun mass production of 192GB SOCAMM2 for NVIDIA Vera Rubin.
The DRAM supply shortage could extend beyond 2027.
Codex received a major update spanning computer use, browser control, and even image generation.
Built a tool that tracks AI agents' memory, decisions, and loops in real time using a 3D graph.
It proposes a recursive development workflow that uses repo documents to prevent context rot in agent work.