Akamai Introduces Agent Sandbox Isolation System to Prevent AI Agent Data Leaks
November 14, 2026 · 1 min read
Also published in Italiano
Akamai has introduced the Akamai Agent Sandbox, a system designed to prevent data leaks from AI agents.
The technology utilizes WebAssembly technology developed with the Cloud Native Computing Foundation to run AI agents within an isolated environment. This mitigation step addresses increased cybersecurity risks caused by autonomous AI agents that can write their own code independently without constant human intervention.
To address financial burdens from high token consumption, Akamai is offering open endpoints compatible with industry standards. This allows companies to shift some workloads to open-weight models for supporting tasks such as code checking, documentation, and unit testing to reduce generative AI costs.
These efficiency strategies include the use of serverless inference computing infrastructure. Akamai aims to allow industries to innovate using autonomous artificial intelligence without security risks or strict dependence on a single service provider.