The Linux Foundation

TechCrunch publishes updated AI glossary

TechCrunch published an updated, regularly maintained AI glossary on 2026-07-03 that explains core AI and generative-AI concepts in plain language. The guide defines terms ranging from AGI, LLMs, chain-of-thought and reinforcement learning to infrastructure topics like compute, token throughput, KV caching and supply-chain concerns dubbed “RAMageddon.” It highlights interoperability developments such as the Model Context Protocol (MCP) — introduced by Anthropic in 2024, handed to the Linux Foundation, and adopted by OpenAI, Google and Microsoft — and discusses model architectures like Mixture of Experts (MoE). The piece is positioned as a living reference for builders, investors and readers trying to keep pace with AI terminology and technical trends.

Context Passing in Multi-Agent AI Systems

Engineering teams face new challenges when capabilities are split across multiple independently deployed AI agents owned by different teams. Microsoft’s Industry Solutions Engineering (ISE) team published a case study describing three evaluated approaches for sharing conversational context across agents: (1) domain agents reading shared storage, (2) making domain agents stateful, and (3) embedding summarized conversation history in each message payload. Microsoft adopted the third approach, sending summarised history inside messages and applying a 10-turn summarisation threshold to balance fidelity and performance. The post contrasts the Model Context Protocol (MCP), which standardises agent-tool connections, with Agent2Agent (A2A), an open peer-to-peer agent communication protocol originally developed by Google and now stewarded via the Linux Foundation. The article highlights governance, security, auditability, and operational benefits of keeping domain agents stateless.

The Harness Trilogy: AI Capability Moves Into the Harness

This analytical synthesis by Gennaro Cuofano (The Business Engineer), published 2026-06-29, argues that AI’s scaling axis has shifted outward from models into the surrounding systems (the “harness”). The author presents three complementary perspectives — industry (Why), personal (Life), and societal (Society) — and argues the same fractal architectural pattern (a principal/authoring core, a swarm of executors, shared memory, and governance gates) repeats at each scale. The essay traces a four-stage migration (pre-training → test-time reasoning → agentic systems → swarm orchestration across 2020–2026), claims organizational and societal forks between adopters and non-adopters, and highlights open problems: replenishing principals (the apprenticeship gap), a governance vacuum, and accelerating cycle times that may outpace adaptation.