It's not just OpenAI models escaping and running riot — experts show how Claude Cowork can break its bonds and access Mac files

Accomplish AI showed Claude Cowork could escape a VM sandbox via Linux zero‑day CVE‑2026‑46331 Agent accessed host Mac files, risking exfiltration of SSH keys, cloud credentials, and more Anthropic shifted Cowork to default cloud execution; local users must harden configs to mitigate exposure Recent

<![CDATA[ <article> <ul><li><strong>Accomplish AI showed Claude Cowork could escape a VM sandbox via Linux zero‑day CVE‑2026‑46331</strong></li><li><strong>Agent accessed host Mac files, risking exfiltration of SSH keys, cloud credentials, and more</strong></li><li><strong>Anthropic shifted Cowork to default cloud execution; local users must harden configs to mitigate exposure</strong></li></ul><p>Recent news of <a href="https://www.techradar.com/pro/security/this-one-was-different-from-anything-we-had-handled-before-hugging-face-confirms-it-was-hit-by-cyberattack-powered-by-an-ai-agent">a ChatGPT agent escaping the sandbox and attacking services on the internet</a> raised quite a few eyebrows, but it seems it’s not the only one capable of running wild. Security researchers Accomplish AI are saying they achieved similar results with Anthropic’s Claude Cowork.</p><p>In a new report, the researchers said they ran a local session in a Mac-hosted <a href="https://www.techradar.com/best/best-virtual-machine-software" target="_blank">virtual Linux machine</a> and then observed as the agent broke free of the VM and started reading and writing files on the underlying system.</p><p>“We connected a folder to a fresh Claude Cowork session, sent one short message, and watched the agent escape the sandbox,” Oren Yomtov, principal security researcher at Accomplish AI, told<em> </em><a href="https://thehackernews.com/2026/07/claude-cowork-flaw-could-let-ai-agent.html" target="_blank"><em>The Hacker News</em></a>. “From inside the VM, it reached the host Mac and read and wrote files all over it, far outside the folder we'd connected, with no permission prompt anywhere.”</p><h2 id="defaulting-to-cloud-execution">Defaulting to cloud execution</h2><p>This means that, in theory, the agent can be used to access or exfiltrate anything that’s stored on the Mac’s user account, including SSH keys, cloud credentials, and more. To break out of the sandbox, the agent exploited CVE-2026-46331 ("pedit COW"), a Linux kernel privilege-escalation vulnerability. This flaw, fixed in mid-June this year, was given a severity score of 7.8/10 (high).</p><p>Accomplish AI disclosed these findings with Anthropic, which allegedly acknowledged them but did not issue a direct fix. However, the version of Claude Cowork that was released afterwards defaults to cloud execution which, the publication claims, addresses the issue. Still, users who opt to run the agent locally rather than in the <a href="https://www.techradar.com/best/best-cloud-storage" target="_blank">cloud</a> will remain exposed. </p><p>Mitigations are possible, though. Users should disable unprivileged user namespaces, grant/revoke seccopm permissions, stop modules autoloading, and restrict sharing of the whole host into the VM. </p><p>"Scope it to the folders that were actually connected instead of all of /, or at least mount it read-only, and run coworkd with ProtectSystem=strict in its own mount namespace so it isn't re-execing binaries a session user can poison," Accomplish AI explained. "Then even a full guest-root has nothing to land on, the last two steps of the chain have nowhere to go."</p> </article> ]]>