Docker Rebrands Around AI Agents with 'Run on Trust' and New Sandbox Tools
Docker has completed a brand refresh that places AI agents at the center of its messaging. The familiar whale logo received a new color palette, and the homepage now prominently displays the slogan “Trust Docker for the agents you don’t.” The company’s marketing frames the change as a response to the risks of giving autonomous agents direct terminal access to developer machines and corporate networks.
Director of Marketing Terry Avnaim explained that agents can independently decide which commands to run, which files to modify, and which services to contact. Docker therefore positions its platform as a controlled environment where access to files and the network can be restricted in advance. The phrase “trust” appears more than twenty times in the accompanying blog post, while containers themselves receive only a brief reaffirmation that they will continue to be developed.
Examination of the product menu reveals that three items carry the “New” label: Sandboxes, AI Governance, and MCP Enterprise Gateway. All three were already available or publicly discussed earlier in 2025. Sandboxes launched on 24 September, AI Governance was referenced in July, and an open-source version of the MCP Gateway has long been included in Docker Desktop. The rebranding therefore serves mainly to surface existing capabilities for an AI-focused audience.
Sandboxes execute each agent inside a dedicated microVM that contains its own kernel and a separate Docker Engine. On Linux the hypervisor is KVM; on Windows it is the Windows Hypervisor Platform. This architecture prevents an agent from sharing the host kernel, unlike ordinary containers. Users invoke the sandbox through a dedicated CLI called sbx with commands such as sbx run claude ~/project-a ~/shared-libs:ro or sbx run --clone claude .. The --clone flag creates an isolated copy of the repository so that destructive experiments do not affect the original working tree.
Network egress is governed by explicit allow-list policies. An administrator can add permitted domains with sbx policy allow network registry.npmjs.org. API keys are never exposed inside the sandbox; instead, a proxy on the host substitutes the real token into request headers. On macOS the feature requires Apple silicon and macOS Sonoma 14 or later; on Linux the minimum is Ubuntu 24.04.
AI Governance is an enterprise console that lets platform teams define organization-wide policies for domains, mount points, and allowed MCP tools. Audit events are streamed to Splunk, Datadog, or Grafana. When centralized governance is enabled, local developer overrides are disabled. MCP Enterprise Gateway adds corporate SSO authentication and automatic credential injection for MCP servers; pricing for both products is available only on request.
Hardened Images have been free in their basic form since 17 December 2025. More than one thousand minimal images based on Debian and Alpine are offered under the Apache 2.0 license. Paid tiers add FIPS-validated images, vulnerability-fix SLAs, and extended support.
The strategic pivot coincides with the leadership of CEO Don Johnson, who previously built cloud infrastructure at Oracle. With container runtimes now widely available from multiple vendors, Docker appears to be targeting the emerging security requirements of organizations that run large numbers of AI agents. Community reaction on Hacker News and Lemmy has been mixed, with praise for the sandbox technology itself but skepticism about the rebranding effort.
Related articles
AI Agents Exploit PaperCut Flaw, Compromising 395 Organizations Across 48 Countries
PaperCut released a security bulletin on 27 August 2026. GreyNoise observed the first real-world code execution just four days later, with the operator moving from initial access to execution in under four hours. After the mass campaign launched, 11 organizations were compromised in only 26 seconds. In total, at least 440 PaperCut NG/MF instances belonging to 395 organizations were breached. The incident is detailed in the November report by Google Threat Intelligence Group and Mandiant, which examines the role of AI agents in the attacks. GTIG also discovered an automated exploit_agent system capable of handling exploitation without constant human input. Additional cases involved LLMjacking and unauthorized configuration changes in Claude Code repositories at cryptocurrency firms.
Google Threat Intelligence Group Warns of Fully Autonomous AI Cyberattacks Within Six Months
Google's Threat Intelligence Group (GTIG) has observed the first real-world instances of attackers using multi-agent AI frameworks to autonomously discover vulnerabilities, generate exploits, and compromise targets without human intervention. The discovery came from an exposed server running a framework called exploit_agent that analyzed software update logs from popular web applications to identify and exploit weaknesses. Four previously defensive tools—ARTEX, CAIRN, CyberStrikeAI, and HexStrike AI—have been repurposed for offensive operations, with documented cases targeting French media, U.S. aerospace, and cryptocurrency firms. GTIG projects that low-complexity automated attacks will become common within three months, while sophisticated fully autonomous operations are expected within six months. The report highlights how these systems compress patch windows, overwhelm security teams with AI-generated proof-of-concepts, and turn detailed vendor advisories into attack playbooks. Defenders are urged to minimize public exposure, accelerate patching cycles to weekly, and tightly control API keys to prevent LLMjacking.
OpenAI Rolls Out KYC, Hardware Keys and Tiered Access for Cybersecurity Tasks in GPT-6.2
OpenAI has introduced stricter access controls for GPT-6.2 Astra and Sol models when users attempt cybersecurity-related tasks. Certain legitimate security research and testing activities now trigger automatic blocks unless the user obtains Daybreak trusted access. The program requires paid subscriptions, FIDO2 hardware keys, Advanced Account Security, and identity verification through Persona. Developers report that even after meeting preliminary requirements, access to reduced refusal rates is not guaranteed and may require separate organizational approval. The changes stem from the models reaching Critical capability levels under OpenAI’s Preparedness Framework for dual-use cybersecurity skills. Similar verification processes previously led to account blocks for users of competing services such as Anthropic’s Claude.
Positive Technologies Releases PT Application Inspector Plugins 3.0.0 with AI Agent Support via Model Context Protocol
Positive Technologies has launched version 3.0.0 of its free PT Application Inspector plugins, introducing native support for AI agents through an integrated MCP server. The update enables AI agents to initiate vulnerability scans on entire projects or specific code changes, review findings, and assist developers in remediation directly within supported environments. Integration via the Model Context Protocol connects tools such as OpenCode, Claude Code, and Codex, providing agents with detailed vulnerability data including exploitation conditions. Static analysis remains the core detection method, while the new AI Hooks feature automatically validates modifications made by agents without requiring full project rescans. The How to Fix capability, which generates AI-driven remediation suggestions, is now available in Visual Studio Code in addition to JetBrains IDEs, with support for on-premise models to keep source code internal. Director of AppSec and WAF Products Anton Zhabolenko stated the goal is to embed vulnerability detection and fixing into the code generation workflow itself.