Eric D. Brown, D.Sc.

ericdbrown.bsky.social

20 October 2026 23:48

JPMorgan won't let outside AI agents into customer accounts, and Microsoft says to trust the model least. Both point at one decision: how much damage a wrong or manipulated agent is allowed to do. https://ericbrown.com/the-insider-problem/ #AI #AIStrategy #AgentSecurity #Judgment #Identity

Rost Glukhov

rosgluk.bsky.social

20 October 2026 14:12

Hindsight for Hermes Agent: Self-Hosted Memory with Local LLMs #Hermes #AgentMemory #SelfHosting #llamacpp #Ollama #PostgreSQL #LLM #Docker https://www.glukhov.org/ai-systems/hermes/hindsight-for-hermes-local-llms/

Linux news

linuxtoday.bsky.social

20 October 2026 01:52

AMD GAIA 0.25 Brings Several Improvements For Local AI Agents On Ryzen AI https://www.phoronix.com/news/AMD-GAIA-0.25-Released

Governed Memory for AI Agents

magnemo.ai

20 October 2026 00:18

One agent writes a fact. Another builds on it. Nothing asks if it was right. A builder on X: memory is "more like developer infrastructure." What checks one agent's notes before another trusts them? Magnemo, open source memory for your AI. Install in one paste. magnemo.ai

The New Stack

thenewstack.io

19 October 2026 22:18

Uvi says AI agents that rebuild company context on every query cost eight times more. Its fix is a shared memory layer, tested only in-house.