ChatGPT Sandbox Flaw Lets Attackers Steal Gmail Data Across Accounts via Hidden Channel
This story is only covered by news sources that have yet to be evaluated by the independent media monitoring agencies we use to assess the quality and reliability of news outlets on our platform. Learn more here.12 Articles
12 Articles
An investigation by Check Point revealed that an internal instance of Artifactory could serve as a hidden channel for a ChatGPT session to send instructions and extract data from someone else’s connected accounts. The infrastructure was removed, but the case exposes the risks of entrusting AI agents with simultaneous access to private code, services and files.
ChatGPT Sandbox Flaw Let A Planted Prompt Ship Victim's Gmail Data To Another Account
Check Point Research disclosed Tuesday that a weakness in ChatGPT's code-execution sandbox let data from one user's connected Gmail account be moved into a separate, attacker-controlled ChatGPT account, with no confirmation prompt shown to the victim. The flaw sat not in the model but in the infrastructure beneath it. Containers running individual conversations cannot address one another directly, Check Point found, but every container could rea…
The Shared Clipboard Inside the Sandbox: Cross-Account Data Leakage in ChatGPT
Research by: Alexey Bukhteyev Check Point Research discovered a covert cross-account command channel through which an attacker could use a victim’s ChatGPT session to execute hidden tasks with the tools, data, and connected apps available to that session. The victim could receive a normal answer to their visible request while the attacker’s task was processed
According to an official publication by Check Point Research, in a report published today, a single command within a ChatGPT conversation could allow a cybercriminal to exploit it, while at the same time answering the user's questions. Based on what we understand, the critical element is not that […]
A failure in ChatGPT discovered by Check Point Research revealed a particularly worrying scenario: malicious instructions could transform a
Coverage Details
Bias Distribution
- There is no tracked Bias information for the sources covering this story.
Factuality
To view factuality data please Upgrade to Premium
