Hearing the Whispers: Black-Box Membership Inference Attacks on Finetuned TTS Models

Text-to-Speech TTS foundation models are increasingly fine-tuned on private datasets to synthesize highly personalized voices, introducing severe privacy risks by exposing both biometric identities and sensitive speech content. Existing black-box membership inference attacks MIAs follow a two-sta...

A SoK for SoCs: Reading the TI Leaves on AI for Cyber Threat Intelligence Generation and Sharing

Cyber Threat Intelligence CTI is essential for defending mission-critical infrastructure, yet the process of transforming raw attack evidence into shareable CTI remains fragmented and understudied. We conduct a literature survey of academic papers, organizing the CTI lifecycle into three stages:...

Auditing Anonymous AI Models: A Four-Stage Protocol for Black-Box Identity Verification

The 2025--2026 AI market has seen a wave of stealth releases: frontier models launched anonymously on developer platforms under codenames. For their users, identity determines data-handling terms, supply-chain risk, and capability expectations. No validated methodology exists for black-box identi...

Hidden Threat in Synthetic Data: Covert Targeted Bias Injection through Benign Text

Synthetic data is increasingly used to train large language models LLMs, yet its security implications remain poorly understood. Prior work on subliminal learning suggests that models can inherit behavioral traits from seemingly unrelated training data. In this work, we investigate whether such...

EvoFlint: An Evolutionary Atlas of Multi-Turn LLM Vulnerabilities

Frontier language models that refuse harmful single-turn prompts often comply when the same intent is reached gradually over many turns, making multi-turn attacks one of the least understood failure modes of large language models. Most automated red-teaming methods treat this as a generation...

Same Request, Different Boundary: Evaluating Cybersecurity Assistance across Conversational Contexts

Large Language Models LLMs can solve complex problems, but their misuse in high-risk domains can lead to severe consequences. Model providers therefore restrict assistance for potentially harmful requests. Refusing all cybersecurity requests would therefore harm legitimate users. Providers need a...

Foundation and Multimodal Large Language Models for Face Presentation and Morph Attack Detection

Face recognition systems are increasingly deployed in security-critical applications, yet they remain vulnerable to presentation and morph attacks. Presentation attack detection PAD and morphing attack detection MAD are therefore essential components of trustworthy face biometrics. Despite...

CVE-2026-54766

Vikunja is an open-source self-hosted task management platform. From 0.21.0 until 2.4.0, the project duplication operation in pkg/models/projectduplicate.go allows an authenticated user who can read a source project to place its duplicate beneath an arbitrary target parent project...

AI Doesn’t Mean the End of Mathematics—at Least Not Yet

This essay was written with Kasra Rafi, and originally appeared in The Guardian. Earlier this month, about 40 top mathematicians gathered at OpenAI's offices to discuss the future of their profession. The meeting was off-the-record, but if recent articles by mathematicians are any guide, it was...

CVE-2026-15679

A flaw was found in Hugging Face PyTorch Image Models. A remote attacker could exploit a deserialization of untrusted data vulnerability by enticing a user to visit a malicious page or open a malicious file. This could lead to arbitrary code execution in the context of the current process. The fl...

Adversarial Training of Linear Models under Stealthy Attacks

Predictive models are widely used in many fields, but are vulnerable to false data injection attacks. To address this, detection schemes and adversarial training have been proposed, but such approaches lack guarantees against stealthy attacks. We therefore propose a detector-based switched model,...

Vulnerable Code Search: Transferable Attack for Code Language Models

Reliable code retrieval is crucial for developer productivity and effective code reuse. However, current neural code language models CLMs powering search tools are susceptible to adversarial attacks targeting non-functional textual elements. In this paper, we introduce a programming...

CVE-2026-79774

Winter CMS versions before 1.2.13 contain an incomplete fix for a Twig sandbox escape vulnerability in System\Twig\SecurityPolicy that allows authenticated backend users with template-editing permissions to bypass sandbox restrictions. Attackers can exploit method forwarding through Eloquent...

CVE-2026-76843

The official Flair wheels for 0.15.0 and 0.15.1 still contain flair/models/clustering.py, whose ClusteringModel.load static method returns pickle.loadsjoblib.loadstrmodelfile and so executes arbitrary Python while loading a model file. Loading a model supplied by an attacker therefore runs that...

Malicious Package

Overview sme-rko-finance-front-shared-entity-groups-models is a malicious package. This package contains malicious code, and its content was removed from the official package manager. While this package might be attempting to impersonate a valid organization, there is no connection between that...

CVE-2026-76602 Joomla Extension - fabrikar.com - Unauthenticated SQL injection in ORDER BY in Fabrik < 4.7.2

Joomla Extension - fabrikar.com - Unauthenticated SQL injection in ORDER BY in Fabrik 4.7.2 - The order parameter in list models is used in queries without validation, allowing read SQLi vectors...

CVE-2026-71513

NLTK before 3.10.3 contains a remote code execution vulnerability in AllowlistUnpickler that validates only the pickle module string and not the global name, allowing attackers to resolve dotted names by attribute traversal to callables outside the allowlisted namespace. Attackers can craft...

CVE-2026-76602: Improper Neutralization of Special Elements used in an SQL Command

Joomla Extension - fabrikar.com - Unauthenticated SQL injection in ORDER BY in Fabrik 4.7.2 - The order parameter in list models is used in queries without validation, allowing read SQLi vectors...

EUVD-2026-63459

Hugging Face PyTorch Image Models checkpoint Deserialization of Untrusted Data Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of Hugging Face PyTorch Image Models. User interaction is required to exploit this...

CVE-2026-15679

The vulnerability is in timm's checkpoint parsing code, which deserializes untrusted PyTorch checkpoint data without validation, enabling arbitrary code execution in the current process. Exploitation requires user interaction (visiting a malicious page or opening a malicious file). No affected ve...