#AI智能体
376 related articles
Why Is GPT-6 Causing Such a Stir? From Conversational Tool to Autonomous Agent
Why is GPT-6 (Astral) causing such a buzz? We break down its autonomous agent capabilities, API-free software operation, OS World 2.0 scores, and how close it really is to AGI.
TraeWork vs WorkBuddy: A Hands-On Comparison — How to Choose the Right AI Agent
Hands-on comparison of ByteDance's TraeWork and Tencent's WorkBuddy AI agents — covering cloud sandbox vs. local execution, features, pricing, and a same-prompt game dev test to help you choose.
Getting Started with WorkBuddy: How Tencent's AI Agent Is Reshaping Workplace Productivity
WorkBuddy is Tencent's AI agent desktop tool that reads local files, executes multi-step tasks, and runs on a schedule. Learn how it differs from ChatGPT and how to use it to boost workplace productivity.
The Rogue AI Agent Problem: Can AI Supervising AI Be the Cure?
As AI agents outpace human review capacity in speed, duration, and scale, enterprises face a critical oversight gap. Can AI supervising AI be the fix?
UN Partners with Google to Make Global Data AI Agent-Ready
The UN is partnering with Google to make global development data AI agent-ready, after UNICEF tests showed leading AI models struggle to accurately retrieve key statistics.
The Two Most Valuable Types of AI Agents: Expert Digital Twins and Enterprise Custom Builds
Distilled from real client projects: the two most valuable AI agents are Expert Agents (domain knowledge digital twins) and Enterprise Custom Agents. Both derive value from scarcity that general LLMs can't replicate.
Skillsync: Enabling AI Chat Sessions to Migrate Freely Between Agents
YC W26 startup Skillsync aims to make AI chat sessions portable across agents, solving context fragmentation and vendor lock-in across platforms like ChatGPT, Claude, and Gemini.
AI Plays Pokémon Red: Jev Clears Two Gyms for Under $2
A Reddit user had AI model Jev play Pokémon Red in real time while Opus 5 built the harness on the fly — clearing two gyms for under $2, showcasing low-cost AI agents.
SpawnRipple: An External Social Environment Built for Autonomous AI Agents
SpawnRipple is an external social environment for autonomous AI agents — providing identity, publishing, discovery, interaction, and API without running any agent models.
Text Agent Store: An AI Agent Marketplace You Can Access by Text Message
Text Agent Store is an AI agent marketplace you access by text — no sign-up, download, or setup needed. Explore how SMS becomes the zero-friction AI interface.
ZeroClick: The New Kind of Store That Sells to AI Agents
ZeroClick helps businesses turn APIs and products into storefronts that AI agents can discover and purchase from directly, bundling payments, pricing controls, and transaction tracking.
Comp AI Raises $34M Series A, Bets on Agentic Security Compliance
Comp AI raises $34M Series A led by Roo Capital and Grand Ventures, betting on "continuously agentic" AI to transform compliance from periodic audits into real-time monitoring.
AI Agents Get to Work: So You Can Focus on What You Love
AI agents are taking on real work — the core value is letting people do more of what they love. Explore the human-AI collaboration trend and its real-world challenges.
Preparing 3D Scenes with Agentic AI: A New Paradigm for Physical AI Simulation
NVIDIA shares how Agentic AI can automatically prepare and validate 3D scenes, building simulation-ready digital twins for robots and autonomous driving systems.
AI Agent Learning Roadmap: From Zero to Real-World Projects
How should you learn AI Agent development? This article outlines a clear, job-ready roadmap: break down the essentials with the LLM+Memory+Tools formula, build a knowledge map, go project-driven, and calculate your ROI.
AI Labs Are Rushing to Hire Internal Auditors — But They Should Lock the Front Door First
AI labs are hiring internal auditors to manage rogue agents — but a simpler, more effective solution may be hiding in plain sight: lock the front door first by restricting agent permissions at the source.
Claude Cowork Merges into Claude: Anthropic Bets on General-Purpose AI Agents
Anthropic merges Claude Cowork and chat into a unified Claude, officially positioning it as a general agent with async task execution. New features roll out first to Pro and Max users.
Insuring AI Agents: How AIUC Uses Insurance Mechanisms to Mitigate Agency Risk
AIUC CEO Rune Kvist explains how insurance can underwrite AI agent behavior after a Series A raise, turning AI trust into quantifiable, priceable risk management.
$20/Month vs. $5M/Year: The Fundamental Difference Between Cursor and Blitzy
Cursor at $20/month vs. Blitzy at up to $5M/year: a Grafana 3M-line codebase test reveals how these AI coding tools differ in context, work unit, and production readiness.
Microsoft's Four IQs Explained: Grounding Capabilities That Give AI Agents Context
What are Microsoft's Foundry IQ, Fabric IQ, Work IQ, and Web IQ? This article breaks down how each grounding capability gives AI agents context from knowledge bases, business data, office apps, and the live web.