#AI智能体
380 related articles
NVIDIA Launches AI Agent Security Platform to Contain Rogue Agent Risks
NVIDIA launches a platform to contain rogue AI agents, focusing on behavioral boundaries, real-time monitoring, and prompt injection defense. An analysis of its capabilities and industry impact.
Powered by 34 AI Agents: Allie Miller's 'Claude Walks' Work Method
Open Machine CEO Allie K. Miller shares her 'Claude walks' workflow: running 34 AI agents in parallel while stepping away. What this agent fleet model means for knowledge workers.
Coze for Beginners: A No-Code Guide to Building AI Agents
A beginner's guide to Coze: the no-code AI agent platform. Learn what AI agents are, why Coze is worth learning, and how to set realistic expectations — no coding required.
Getting Started with AI Agent Development: A Practical Breakdown of the Four Core Modules
A beginner's guide to AI Agent development: break down the four core modules — prompt architecture, tool calling, RAG, and ReAct workflows — to fix hallucinations, loops, and unreliable output.
When AI Agents Are Untrusted: How Should Payment Authorization Be Designed?
When AI agents can spend money autonomously, how should payment authorization be designed? This article examines x402Shield's model: separating intent from approval, exact intent binding, and durable authorization state.
From CRUD Developer to AI Agent Builder: A Pragmatic Learning Roadmap
A Reddit-validated AI Agent learning roadmap for CRUD developers: covers Wrapper vs. Workflow vs. Agent, the agent loop, tool calling, RAG, evals, guardrails, security, and how to see through multi-agent demo hype.
Microsoft WorkIQ Deep Dive: Injecting Work Context into AI Agents
Microsoft WorkIQ provides an intelligent context layer for AI Agents built on Microsoft 365 data, with REST API, A2A, and MCP integration. This deep dive covers WorkIQ vs. Graph, permissions, open-source WorkIQ OS, and Margo — an Agent managing 250K emails.
Praxa: A Governance Framework for AI Agent Execution with Traceable Chains from Proposal to Verification
arXiv paper proposes Praxa, an AI agent execution harness that makes the authorization-to-effect chain explicit and testable via deterministic admission, brokered execution, and external read-back — with four honest evidence chains and clear capability boundaries.
OpenAI DevDay 2026 Full Breakdown: GPT-4.1 So and the Always-On Codex Agent
OpenAI DevDay 2026 unveiled GPT-4.1 So and the always-on Codex agent. A plain-language breakdown of the new models, ultra-fast mode, cloud coding, and how Codex compares to Manus and Muse.
NVIDIA DOCA Agent Skills: Accelerating BlueField Application Development
NVIDIA DOCA Agent Skills injects BlueField DPU domain knowledge into AI agents, bridging the gap general-purpose models face in specialized infrastructure development.
Shopify Launches Canvas: Build Your Online Store Through Conversation with AI
Shopify's new Canvas tool lets merchants build and customize online stores by chatting with AI agent Sidekick, with real-time previews of every change.
Brian Chesky: AI Agents Need Their Own Operating System
Airbnb CEO Brian Chesky argues AI agents need a dedicated AI-native OS, and discusses making platforms agent-friendly and the challenges of consumer AI adoption.
OpenAI Launches Dots: Always-On AI Agents Come to ChatGPT
OpenAI launches Dots, always-on AI agents powered by GPT-6 Astra with dedicated cloud computers, 4000+ app integrations, and 24/7 task execution for ChatGPT Pro and Business Premium users.
Photon Raises $4.5M to Replace Mobile Apps with AI Agents
AI startup Photon raises $4.5M to help developers build AI agents on iMessage, SMS/RCS, and email — betting consumers will ditch traditional mobile apps.
One Kimi Hidden Among Four Claudes: Can AI Sniff Out an Undercover Model?
A creative AI experiment: one Kimi hidden among four Claudes. Claude used behavioral fingerprinting and SHA-256 hash commitments to expose the mole — while Kimi's reverse attempt failed.
Claude Opus 5.5 Levels Up, But Meta Muse Is the Real Game-Changer
Anthropic's Opus 5.5 upgrades writing and frontend design. OpenAI adds live tool calls to GPT voice. Meta Muse tops the App Store for free. Deep dive into the converging personal agent race.
Sam Altman Interview: AI Agent Collaboration and the New Paradigm for Developer Platforms
Sam Altman at OpenAI Dev Day discusses Dots agents, Space live documents, open developer platform strategy, and democratizing AI speed with custom chips.
SCLATE: A Unified Training and Evaluation Substrate for Continual-Learning Agents
SCLATE is a unified execution substrate for continual-learning agents that coordinates session events, cron jobs, and memory consolidation via an open event scheduler and adapter mechanism.
Flow Engineering Valued at $750M: AI Agents Enter Hardware Design
Flow Engineering raises at a $750M valuation with backing from Valor, Atreides, and Sequoia. Roelof Botha joins the board as an angel investor as the startup brings AI agents to hardware design.
OpenAI Decisions API: How Fast, Cheap Intelligence Tames Swarms of AI Agents
OpenAI launches Decisions API as a lightweight arbiter for swarms of AI agents, confirming speed and cost as the next frontier of AI competition. Here's what it means.