The-Decoder Google Launches Agent Compute Engine with Isolated MicroVMs for AI Agents
Article Content
- •Google's Agent Compute Engine provides isolated microVMs for AI agents.
- •The service aims to improve efficiency by allowing agents to pause when idle.
- •Governance controls include token tracking and support for third-party identity providers.
Google has announced the Agent Compute Engine, a service that provides each AI agent with its own isolated micro virtual machine (microVM). This service, which powers the Gemini AI agents, is set to open for preview later this year. Each microVM is secured by Google's Titanium chip and can pause when idle, allowing for efficient resource management. Google claims that this setup can support ten times as many agents on a single server, with a resume time of less than 500 milliseconds. The service also includes governance features such as token tracking, model routing, and support for third-party identity providers. The announcement was made on October 22, 2026, and no security vulnerabilities or exploitation concerns were reported in relation to this new service.
Ask AI about this cluster
Answers cite the sources they use
Timeline
More articles in this cluster (5)
Following this threat?
Track CVE-2025-32711 in your own feed — you're alerted when they show up in new reporting, leak sites or exploitation.
Free account · no card needed
Common questions
What is the Agent Compute Engine?
When will the Agent Compute Engine be available for preview?
What governance features does the service include?
Continue Reading
Enterprises Struggle with AI Governance Amid Rising Shadow AI Usage As AI adoption accelerates, only 18% of enterprises maintain a complete AI inventory, according to IBM. Netskope found that 47% of enterprise GenAI users are still utilizing personal AI applications, leading to significant visibility gaps for security teams. The rapid spread of shadow AI complicates governance, as…
Critical BeyondTrust 9.3 Vulnerability Actively Exploited by Threat Actor A critical vulnerability (CVE-2026-102764) in BeyondTrust Remote Support has been discovered and is actively being exploited by the threat group ARMORY CASTLE, reportedly linked to China. The vulnerability, which has a CVSS score of 9.3, allows unauthenticated attackers to upload files and potentially compromise host…