SesameOp: Novel backdoor uses OpenAI Assistants API for command and control

Microsoft Incident Response – Detection and Response Team DART researchers uncovered a new backdoor that is notable for its novel use of the OpenAI Assistants Application Programming Interface API as a mechanism for command-and-control C2 communications. Instead of relying on more traditional...

Would you sext ChatGPT? (Lock and Code S06E22)

This week on the Lock and Code podcast … In the final, cold winter months of the year, ChatGPT could be heating up. On October 14, OpenAI CEO Sam Altman said that the "restrictions" that his company previously placed on their flagship product, ChatGPT, would be removed, allowing, perhaps, for...

Arbitrary File Read

flowise is vulnerable to an arbitrary file read. The vulnerability is due to improper validation of the chatId parameter in the /api/v1/get-upload-file and /api/v1/openai-assistants-file/download endpoints, which allows an attacker to read unintended files on the local filesystem and potentially...

Atlas browser’s Omnibox opens up new privacy and security risks

It seems that with every new agentic browser we discover yet another way to abuse one. OpenAI recently introduced a ChatGPT based AI browser called Atlas. It didn’t take researchers long to find that the combined search and prompt bar—called the Omnibox—can be exploited. By pasting a specially...

Build AI Agents with Spin and the OpenAI Agents SDK

...

NSFW ChatGPT? OpenAI plans “grown-up mode” for verified adults

If you've had your fill of philosophical discussions with ChatGPT, CEO Sam Altman has news for you: the service will soon be able to engage in far less highbrow conversations of the sexual kind. That's right—sexting is coming to ChatGPT. Are we really surprised? It marks a change in sentiment for...

‘ChatGPT Tainted Memories’ Exploit Enables Command Injection in Atlas Browser

LayerX Security found a flaw in OpenAI’s ChatGPT Atlas browser that lets attackers inject commands into its memory, posing major security and phishing risks...

ChatGPT Atlas Browser Can Be Tricked by Fake URLs into Executing Hidden Commands

The newly released OpenAI ChatGPT Atlas web browser has been found to be susceptible to a prompt injection attack where its omnibox can be jailbroken by disguising a malicious prompt as a seemingly harmless URL to visit. "The omnibox combined address/search bar interprets input either as a URL to...

Amazon Explains How Its AWS Outage Took Down the Web

Plus: The Jaguar Land Rover hack sets an expensive new record, OpenAI’s new Atlas browser raises security fears, Starlink cuts off scam compounds, and more...

CVE-2025-60511

Moodle OpenAI Chat Block plugin 3.0.1 2025021700 suffers from an Insecure Direct Object Reference IDOR vulnerability due to insufficient validation of the blockId parameter in /blocks/openaichat/api/completion.php. An authenticated student can impersonate another user's block e.g., administrator...

CVE-2025-60511

Moodle OpenAI Chat Block plugin 3.0.1 2025021700 suffers from an Insecure Direct Object Reference IDOR vulnerability due to insufficient validation of the blockId parameter in /blocks/openaichat/api/completion.php. An authenticated student can impersonate another user's block e.g., administrator...

Evaluating Large Language Models in Detecting Secrets in Android Apps

Mobile apps often embed authentication secrets, such as API keys, tokens, and client IDs, to integrate with cloud services. However, developers often hardcode these credentials into Android apps, exposing them to extraction through reverse engineering. Once compromised, adversaries can exploit...

EUVD-2025-35199

Moodle OpenAI Chat Block plugin 3.0.1 2025021700 suffers from an Insecure Direct Object Reference IDOR vulnerability due to insufficient validation of the blockId parameter in /blocks/openaichat/api/completion.php. An authenticated student can impersonate another user's block e.g., administrator...

CVE-2025-60511

Moodle OpenAI Chat Block plugin 3.0.1 2025021700 suffers from an Insecure Direct Object Reference IDOR vulnerability due to insufficient validation of the blockId parameter in /blocks/openaichat/api/completion.php. An authenticated student can impersonate another user's block e.g., administrator...

Moodle OpenAI Chat Block plugin security vulnerability

Moodle OpenAI Chat Block plugin is a large model chat plugin for Moodle open source. A security vulnerability exists in version 3.0.1 of the Moodle OpenAI Chat Block plugin, which stems from insufficient validation of the blockId parameter and could lead to an insecure direct object reference...

CVE-2025-60511

Moodle OpenAI Chat Block plugin 3.0.1 2025021700 suffers from an Insecure Direct Object Reference IDOR vulnerability due to insufficient validation of the blockId parameter in /blocks/openaichat/api/completion.php. An authenticated student can impersonate another user's block e.g., administrator...

CVE-2025-60511

The CVE-2025-60511 instance affects Moodle OpenAI Chat Block plugin 3.0.1, with an Insecure Direct Object Reference (IDOR) due to insufficient validation of the blockId parameter in /blocks/openai_chat/api/completion.php. An authenticated student can impersonate another user’s block (e.g., admini...

A week in security (October 13 – October 19)

Last week on Malwarebytes Labs: Prosper data breach puts 17 million people at risk of identity theft Under the engineering hood: Why Malwarebytes chose WordPress as its CMS Video call app Huddle01 exposed 600K+ user logs Mango discloses data breach at third-party provider Roku accused of selling...

Over 100 VS Code Extensions Exposed Developers to Hidden Supply Chain Risks

New research has uncovered that publishers of over 100 Visual Studio Code VS Code extensions leaked access tokens that could be exploited by bad actors to update the extensions, posing a critical software supply chain risk. "A leaked VS Code Marketplace or Open VSX PAT personal access token allow...

Researchers break OpenAI guardrails

The maker of ChatGPT released a toolkit to help protect its AI from attack earlier this month. Almost immediately, someone broke it. On October 6, OpenAI ran an event called DevDay where it unveiled a raft of new tools and services for software programmers who use its products. As part of that, i...