ThreatExchange

Projects in this Repository This repository originally started as code to support Meta's ThreatExchange API, but over time has grown to include a number of projects to support signal exchange and content moderation in general. Below are a list of sub-projects. PDQ Image Hashing and Similarity...

CVE-2026-81161

Privilege Defined With Unsafe Actions vulnerability in Drupal Content Moderation Notifications allows Privilege Escalation. This issue affects Content Moderation Notifications versions: from 0.0.0 to 3.9.0...

EUVD-2026-70004

Privilege Defined With Unsafe Actions vulnerability in Drupal Content Moderation Notifications allows Privilege Escalation. This issue affects Content Moderation Notifications versions: from 0.0.0 to 3.9.0...

CVE-2026-81161

Content Moderation Notifications for Drupal (versions 0.0.0 through 3.9.0) contains a privilege escalation vulnerability caused by a permission to configure email templates containing Twig code that was not marked as restricted . A site administrator may inadvertently grant this permission to les...

Content Moderation Notifications - Moderately critical - Access bypass - SA-CONTRIB-2026-107

The module provides a permission that allows users to configure email templates containing Twig code. This permission was not marked as restricted. A site administrator might inadvertently grant this permission to less-trusted users. This would allow those users to execute Twig within email...

GHSA-6C4R-FMH3-7RH8 vLLM: Processing differential in multi-channel audio downmixing enables hidden-input/moderation bypass for audio models

Issue Description Librosa defaults to using numpy.mean for mono downmixing tomono, while the international standard ITU-R BS.775-4 specifies a weighted downmixing algorithm. This discrepancy results in: - Inconsistency between audio heard by humans e.g., through headphones/regular speakers and...

PYSEC-2026-2485 FastChat has a Content Moderation Bypass via Arena Side-by-Side Views

A vulnerability was detected in lm-sys fastchat up to 0.2.36. Impacted is the function addtext of the component Arena Side-by-Side View Handler. The manipulation results in incorrect control flow. The attack can be launched remotely. The exploit is now public and may be used. The root cause was...

FastChat has a Content Moderation Bypass via Arena Side-by-Side Views

A vulnerability was detected in lm-sys fastchat up to 0.2.36. Impacted is the function addtext of the component Arena Side-by-Side View Handler. The manipulation results in incorrect control flow. The attack can be launched remotely. The exploit is now public and may be used. The root cause was...

DRUPAL-CONTRIB-2026-039

This module configures default editorial workflows for LocalGov Drupal content types. It provides a Drupal content moderation workflow, a content approvals dashboard, content scheduling and content preview. The module doesn't sufficiently restrict access to a view of Service Contacts at which...

LocalGov Workflows - Moderately critical - Information disclosure - SA-CONTRIB-2026-039

This module configures default editorial workflows for LocalGov Drupal content types. It provides a Drupal content moderation workflow, a content approvals dashboard, content scheduling and content preview. The module doesn't sufficiently restrict access to a view of Service Contacts at which...

PT-2026-46112

Name of the Vulnerable Software and Affected Versions LocalGov Workflows versions 0.0.0 through 1.6.0 Description Missing authorization allows forceful browsing within the module, which configures default editorial workflows for content types, including content moderation, approvals dashboards,...

GHSA-F3Q6-69F3-VWCH FastChat has a Content Moderation Bypass via Arena Side-by-Side Views

A vulnerability was detected in lm-sys fastchat up to 0.2.36. Impacted is the function addtext of the component Arena Side-by-Side View Handler. The manipulation results in incorrect control flow. The attack can be launched remotely. The exploit is now public and may be used. The root cause was...

FastChat has a Content Moderation Bypass via Arena Side-by-Side Views

A vulnerability was detected in lm-sys fastchat up to 0.2.36. Impacted is the function addtext of the component Arena Side-by-Side View Handler. The manipulation results in incorrect control flow. The attack can be launched remotely. The exploit is now public and may be used. The root cause was...

DRUPAL-CONTRIB-2026-006

This Drupal Canvas module is a new visual page builder for Drupal. You can create reusable components that match your design system, drag them onto a page, edit content in place, preview changes across multiple pages, and undo mistakes with ease. The module doesn't sufficiently validate access to...

PT-2026-5242

Name of the Vulnerable Software and Affected Versions Drupal Canvas versions prior to 1.0.4 Description The Drupal Canvas module has an authorization issue that allows forceful browsing of Canvas Pages when they are unpublished. The module does not adequately validate access to Canvas Pages,...

TikTok Forms U.S. Joint Venture to Continue Operations Under 2025 Executive Order

TikTok on Friday officially announced that it formed a joint venture that will allow the hugely popular video-sharing application to continue operating in the U.S. The new venture, named TikTok USDS Joint Venture LLC, has been established in compliance with the Executive Order signed by U.S...

Grok apologizes for creating image of young girls in “sexualized attire”

Another AI system designed to be powerful and engaging ends up illustrating how guardrails routinely fail when development speed and feature races outrun safety controls. In a post on X, AI chatbot Grok confirmed that it generated an image of young girls in “sexualized attire.” The potential...

Jailbreaking Attacks Vs. Content Safety Filters: How Far Are We in the LLM Safety Arms Race?

As large language models LLMs are increasingly deployed, ensuring their safe use is paramount. Jailbreaking, adversarial prompts that bypass model alignment to trigger harmful outputs, present significant risks, with existing studies reporting high success rates in evading common LLMs. However,...

CVE-2025-65113

ClipBucket v5 contains an authorization bypass in the AJAX flagging system that allows any unauthenticated user to flag content (users, videos, photos, collections). Affected versions are prior to 5.5.2; this issue can enable mass flagging and moderation abuse. The vulnerability has been patched ...

EUVD-2025-199885

ClipBucket v5 is an open source video sharing platform. Prior to version 5.5.2 - 164, an authorization bypass vulnerability in the AJAX flagging system allows any unauthenticated user to flag any content users, videos, photos, collections on the platform. This can lead to mass flagging attacks,...