CVE-2026-6596

A security flaw has been discovered in langflow-ai langflow up to 1.1.0. This issue affects the function createuploadfile of the file src/backend/base/Langflow/api/v1/endpoints.py of the component API Endpoint. The manipulation results in unrestricted upload. It is possible to launch the attack...

PT-2026-33831

Name of the Vulnerable Software and Affected Versions Flowsint affected versions not specified Description Flowsint is an open-source OSINT graph exploration tool used for cybersecurity investigation, transparency, and verification. A remote attacker can create a sketch and trigger the org to asn...

Explainable Attention-Based LSTM Framework for Early Detection of AI-Assisted Ransomware Via File System Behavioral Analysis

Ransomware continues to evolve as one of the most disruptive cyber threats, with recent variants increasingly leveraging automated and AI-assisted techniques to evade traditional signature-based defenses. Early detection of such attacks remains a significant challenge, particularly when malicious...

$13.74M Hack Shuts Down Sanctioned Grinex Exchange After Intelligence Claims

Grinex, a Kyrgyzstan-incorporated cryptocurrency exchange sanctioned by the U.K. and the U.S. last year, said it's suspending operations after it blamed Western intelligence agencies for a $13.74 million hack. The exchange said it fell victim to what it described as a large-scale cyber attack tha...

Server-side Request Forgery (SSRF)

Overview @nocobase/plugin-ai is a Create AI employees with diverse skills to collaborate with humans, build systems, and handle business operations. Affected versions of this package are vulnerable to Server-side Request Forgery SSRF via the workflow HTTP request and custom request plugins, which...

EUVD-2025-209469

An access control vulnerability was discovered in the Threat Intelligence functionality due to a specific access restriction not being properly enforced for users with view-only privileges. An authenticated user with view-only privileges for the Threat Intelligence functionality can perform...

Important: Red Hat Security Advisory: Red Hat AI Inference Server 3.3.1 (ROCm)

Red Hat AI Inference Server 3.3.1 ROCm is now available. Red Hat® AI Inference Server...

poker-exploit-ai

No d...

OpenHarness 安全漏洞

OpenHarness is a lightweight development and runtime framework for Data Intelligence Lab@HKU, open source in nature. There is a security vulnerability in OpenHarness; this vulnerability stems from incomplete path normalization in the permission checker, which may lead to access to sensitive files...

Important: Red Hat Security Advisory: RHTAS 1.4 - GA Release of Model Transparency

The GA release of the RHTAS Model Transparency CLI image. For more details please visit the product documentation at https://access.redhat.com/documentation/en-us/redhattrustedartifactsigner/1.4 The RHTAS Model Transparency CLI image can be used to sign and verify AI/ML workloads...

A Bootiful Podcast: the legendary Craig Walls

Hi Spring fans! In this installment we talk to the legendary Craig Walls, author of Spring In Action , Spring AI in Action , and more!...

Too Private to Tell: Practical Token Theft Attacks on Apple Intelligence

Apple Intelligence is a generative AI GenAI service provided by Apple on its devices. While offering a similar set of features as other similar GenAI services, Apple Intelligence is claimed to be designed with an extra focus on user security and privacy through a two-stage authentication and...

AI clickbait can turn your notifications into a scam feed

Pushpaganda is the name researchers have given to an AI-assisted ad fraud, social engineering, and scareware operation targeting mobile users. For most people, Pushpaganda starts as something that looks completely normal. For example, a recommended article in your Google Discover feed the...

Incident response for AI: Same fire, different fuel

In this article 1. The fundamentals still hold 2. Where AI changes the equation 3. Closing the gaps in telemetry, tooling, and response 4. The human dimension 5. Looking ahead When a traditional security incident hits, responders replay what happened. They trace a known code path, find the defect...

Incident response for AI: Same fire, different fuel

In this article 1. The fundamentals still hold 2. Where AI changes the equation 3. Closing the gaps in telemetry, tooling, and response 4. The human dimension 5. Looking ahead When a traditional security incident hits, responders replay what happened. They trace a known code path, find the defect...

Deterministic + Agentic AI: The Architecture Exposure Validation Requires

Few technologies have moved from experimentation to boardroom mandate as quickly as AI. Across industries, leadership teams have embraced its broader potential, and boards, investors, and executives are already pushing organizations to adopt it across operational and security functions. Pentera’s...

The Deepfake Nudes Crisis in Schools Is Much Worse Than You Thought

An analysis by WIRED and Indicator found nearly 90 schools and 600 students around the world impacted by AI-generated deepfake nude images—and the problem shows no signs of going away...

CVE-2025-40897

An access control vulnerability was discovered in the Threat Intelligence functionality due to a specific access restriction not being properly enforced for users with view-only privileges. An authenticated user with view-only privileges for the Threat Intelligence functionality can perform...

CVE-2025-40897

An access control vulnerability was discovered in the Threat Intelligence functionality due to a specific access restriction not being properly enforced for users with view-only privileges. An authenticated user with view-only privileges for the Threat Intelligence functionality can perform...

CVE-2025-40897 Incorrect authorization for Threat Intelligence in Guardian/CMC before 26.0.0

An access control vulnerability was discovered in the Threat Intelligence functionality due to a specific access restriction not being properly enforced for users with view-only privileges. An authenticated user with view-only privileges for the Threat Intelligence functionality can perform...