diff --git a/topics/ai-sandbox/ai-sandbox.png b/topics/ai-sandbox/ai-sandbox.png
new file mode 100644
index 000000000000..d981e655563f
Binary files /dev/null and b/topics/ai-sandbox/ai-sandbox.png differ
diff --git a/topics/ai-sandbox/index.md b/topics/ai-sandbox/index.md
new file mode 100644
index 000000000000..f3952af69738
--- /dev/null
+++ b/topics/ai-sandbox/index.md
@@ -0,0 +1,9 @@
+---
+aliases: ai-sandboxing, llm-sandbox
+display_name: AI Sandbox
+logo: ai-sandbox.png
+short_description: An isolated environment for safely running, testing, and evaluating AI-generated code and autonomous agents.
+topic: ai-sandbox
+related: sandbox, llm, ai-agents, code-execution, containerization
+---
+An AI sandbox is a secure, isolated runtime environment designed to execute AI-generated code, run autonomous agents, and evaluate large language model (LLM) outputs without risk to the host system. Sandboxes prevent untrusted code from accessing the network, filesystem, or other sensitive resources. Common approaches include container-based isolation (Docker, gVisor), WebAssembly runtimes, and MicroVM hypervisors. AI sandboxes are foundational to agentic AI systems where LLMs write and execute code on behalf of users.