Microsoft Launches Multi-Agent AI System Magentic-One

On November 4, Microsoft announced “Magentic-One,” an AI system composed of five agents. This system can plan and execute complex multi-step tasks, such as writing code or ordering food, offering capabilities beyond traditional generative AI chatbots.

Magentic-One consists of the following five agents:

  • Orchestrator: The lead agent that divides tasks and manages progress.
  • WebSurfer: An agent capable of operating a browser.
  • FileSurfer: Capable of reading local files.
  • Coder: Performs code creation and information analysis.
  • ComputerTerminal: Provides access to a console shell.

The Orchestrator creates a “Task Ledger” to plan tasks and uses a “Progress Ledger” to confirm task completion and track progress. In tests using the evaluation tool AutoGenBench, Magentic-One showed results surpassing GPT-4 but did not reach human-level accuracy. Microsoft warns that Magentic-One operates in a digital world designed for humans and may cause irreversible consequences.

Microsoft

  • Related Posts

    Replit Agent’s Rampage Can Wipe Out Days of Work! – Techniques to Prevent Such Tragedy

    Replit Agent is a revolutionary AI programming partner that builds development environments, writes code, and even deploys applications—all based on natural language instructions. With just a few sentences, it can instantly transform your ideas into a functioning app, offering a…

    Understanding Anthropic’s MCP: The Future of AI Communication Protocols

    In this blog, we delve into the transformative potential of Anthropic’s Model Context Protocol (MCP), a groundbreaking framework designed to standardize communication between large language models and their surrounding environments. Join us as we explore the reasons behind its acclaim…

    Leave a Reply

    Your email address will not be published. Required fields are marked *

    You Missed

    Stanford University’s 2025 AI Index Report – Summary of Key Findings

    Stanford University’s 2025 AI Index Report – Summary of Key Findings

    Replit Agent’s Rampage Can Wipe Out Days of Work! – Techniques to Prevent Such Tragedy

    Replit Agent’s Rampage Can Wipe Out Days of Work! – Techniques to Prevent Such Tragedy

    Evolution of AI Models (Jan–Mar 2025)

    Evolution of AI Models (Jan–Mar 2025)

    Exploring DeepSeek: The Future of Inference Learning through Reinforcement Learning

    Exploring DeepSeek: The Future of Inference Learning through Reinforcement Learning

    Generative AI and Dark Patterns in UX Design

    Generative AI and Dark Patterns in UX Design

    Understanding Anthropic’s MCP: The Future of AI Communication Protocols

    Understanding Anthropic’s MCP: The Future of AI Communication Protocols