Top news
Investing in the public benefit OfficialAnthropic NewsOct 18Popular Musk Pledges SpaceX Shares to ‘Every Kid in America’ WebBloomberg TechnologyOct 19Popular Websites are closing their doors to AI agents WebThe Verge AIOct 17 This is the new ChatGPT Voice, powered by GPT Live OfficialOpenAI on YouTubeOct 18 How AI giants will reshape global philanthropy WebAxios AIOct 19 Gemini 4 Argon Really Surprised Me WebTwo Minute PapersOct 19 Google Gemini 4 agents just beat Minecraft with no human intervention WebEngadget AIOct 19 Who could it be? 🧐 OfficialGoogle on YouTubeOct 18 Anthropic gives outside AI watchdog “employee-like” access WebArs Technica AIOct 18 What are self-replicating ‘AI worms’, and why is Anthropic worried about them? WebThe Guardian AIOct 18
GlossaryAgentsIntermediate

Agentic

Also written: agentic AI systems, agentic workflows, Agentic misbehavior, Agentically, Agentic technology, Autonomous AI, Agentic knowledge work, Autonomous knowledge work

AI systems that can act autonomously, plan, and use tools to achieve goals.

In plain terms

Agentic AI refers to artificial intelligence systems designed to operate independently. Instead of just responding to a direct command, these systems can figure out the steps needed to reach a goal, make a plan, and even use tools to get the job done.

Think of an AI assistant that can book your travel. It wouldn't just ask for dates. It would understand your request, search for flights and hotels, compare options, and make the bookings, all on its own. This involves planning, decision-making, and interacting with other software or services.

The expert version

Agentic AI systems are designed for autonomous operation, capable of planning, reasoning, and executing actions to achieve predefined objectives. These systems often leverage tool use, allowing them to interact with external environments or software to gather information or perform tasks. The core capability involves breaking down complex goals into a sequence of actionable steps and dynamically adapting the plan based on intermediate results or environmental feedback.

Tradeoffs include increased complexity in design and safety considerations, as autonomous action requires robust error handling and alignment mechanisms. Current research focuses on improving planning capabilities, tool integration, and the ability to handle more complex, multi-step tasks with greater reliability.

Why it matters now

Agentic AI systems are a major focus as companies like Google and Microsoft are integrating them into their core products, such as coding assistants and robotics. This trend shows a shift towards AI that can operate more independently, making decisions and executing tasks without constant human input, as seen in Google's Gemini Notebook and Microsoft's Copilot updates.

Where it shows up 136 items, newest first
How each company uses it
Anthropic Claude
Anthropic is exploring agentic workflows by enabling AI agents to easily use specialized tools for tasks like customer support.
Google Gemini
Google is heavily investing in agentic AI, developing new Gemini models and resources to build AI agents that can reason, output files, and control robots.
Microsoft Copilot
Microsoft is making agentic AI a reality in its products by enabling Copilot to automatically fix code security alerts and integrating advanced AI models for coding.
OpenAI ChatGPT
OpenAI is observing the growing adoption of agentic AI in European businesses, indicating a global trend towards more autonomous AI applications.
Related terms

Autonomous agentsTool usePlanningReasoningLLM orchestrationAI workflows

Often confused with
AI Assistants
AI assistants typically respond to direct commands, while agentic
Robotics
Robotics involves physical embodiment and action in the real world, wh
Expert Systems
Expert systems rely on predefined rules for decision-making, whi
Origin

The concept of agentic AI has roots in artificial intelligence research, particularly in areas like intelligent agents, planning, and multi-agent systems. The term has gained prominence with the rise of large language models capable of more complex reasoning and action.

Try this about 5 minutes

Use a tool-use enabled AI like ChatGPT with plugins or a Gemini Advanced interface. Ask it to plan a weekend trip to a specific city, including booking hypothetical flights and hotels. Observe how it breaks down the request, identifies necessary steps, and simulates using tools to find information.

Try it: Playbook entries that use thisPlaybook →

The Playbook: reusable how-tos, prompts, configs and tips pulled out of the news.

Questions people ask

What's the difference between a regular AI and an agentic AI?
Regular AI typically responds to direct prompts. Agentic AI can take a goal, plan the steps to achieve it, and execute those steps autonomously, often using tools.
Can agentic AI make mistakes?
Yes, agentic AI systems can make mistakes. Their autonomous nature means they need robust error checking and safety mechanisms to ensure they act reliably and align with intended goals.

First seen on Subvolts Dec 14, 2023. Updated Oct 18, 2026. Written by an AI model from the items above and reviewed against them weekly; report an error.