JevAdvBench: A Benchmark and Black-Box Attacks for Reinforcement Learning for Calibrated Decisions Models

Models trained with reinforcement learning for calibrated decisions RLCD, such as Jev, answer a typed question about an input, the state, with a probability, a choice, or a score, and software acts on the answer without a person reading it. Their robustness has not been measured: adversarial...

FeatMark: Feature-Level Watermark Protection against Mimicry Attacks with Diffusion Models

Text-to-image diffusion models enable data-efficient "mimicry" attacks, wherein adversaries fine-tune the model on a handful of public photos to synthesize convincing forgeries of a target individual. A common countermeasure is to embed imperceptible, low-energy watermarks, yet recent studies sho...

TP-CRIV: A Framework for Third-Party Challenge-Response Identity Verification of AI Models

Artificial intelligence AI models are increasingly deployed through remote services, making model misappropriation a growing concern. Existing approaches, including watermarking, fingerprinting, and model similarity analysis, primarily rely on predefined evidence or direct behavioral comparison a...

Trusted Model Environment for Private Semantic Computations

A private semantic computation primitive enables parties to privately compute over structured and unstructured data that requires understanding its semantics, context, and relationships. Standard cryptographic primitives e.g., multiparty computation do not readily support such computation...

This Windows Malware is Built to Let Up to Four AI Models Vote on Its Next Move

A Windows malware called CLOSEDQUORUM is built to take orders from a vote of up to four AI models instead of an attacker's server, Cisco Talos said on September 22. The models can choose to steal Windows credentials, saved browser passwords, and crypto wallet data. Talos has not seen this setup...

Your Model Is Leaking: Covert Information Transfer through LLM Residual Streams

Privacy-sensitive organizations may run large language models LLMs in restricted or air-gapped environments while exporting selected diagnostic artifacts. We show that a compromised runtime component can hide sensitive information in intermediate activations that are allowed to leave the restrict...

PT-2026-97127

Name of the Vulnerable Software and Affected Versions recommenders-team recommenders versions prior to 1.2.2 Description A flaw in the Dict Loading component allows remote attackers to trigger deserialization. This occurs through the pickle.load function within the...

A Bulletproof Business? Towards Detecting Infrastructure-As-A-Service Offerings on Telegram

Cybercriminal operations increasingly depend on reusable digital infrastructure---including hosting, proxies, and virtual private networks VPNs---rented through Cybercrime-as-a-Service markets and advertised on platforms such as Telegram. We present a taxonomy for identifying Telegram messages...

Unlocking Cross-Scenario Physical Layer Security: A Mixture-Of-Experts Framework with Generative Diffusion Models

The future 6G networks are expected to incorporate a proliferation of wireless services in diverse environments, which presents a significant challenge for information security. Conventionally optimization always requires recalculation and learning strategy often suffers poor generalization, whic...

Privacy Leakage through AI-Mediated Analysis of Smartphone Data

Over the past thirty years, the online advertising industry built a large-scale data collection ecosystem, with the goal of tracking a user's online activity to infer their demographics and interests. Traditionally, the ecosystem relied upon the collation and analysis of highly-structured text da...

On the Security and Privacy of LLMs in Mobility

The mobility sector is undergoing a paradigm shift driven by advances in Generative Artificial Intelligence. With a global market valued at approximately 2.9 trillion dollars annually, considering only cars, the integration of these technologies has the potential to impact more than 1.5 billion...

Decoding the Legalese: A Scalable and Quantitative Framework for Analyzing Corporate Privacy Policies

Even though privacy policies are the primary mechanism organizations use to disclose how they collect, process, and share personal data, they are difficult for average users to interpret, perhaps by design, due to their verbosity and dense legal language. Importantly, there is a lack of...

SSP-Bench: A Hybrid Data Generation Framework for Safety, Security, and Privacy Evaluation

Evaluation of large language models LLMs for safety, security, and privacy SSP relies heavily on static benchmarks, which suffer from score saturation, data contamination, and aggregation artifacts, and fail to capture sensitivity to linguistic variation. As a result, models that perform well on...

CVE-2026-93764

Mongoid may omit encryption rules for fields declared on embedded models when generating the client-side field-level encryption schema. Applications that enable this feature can therefore store values intended to be encrypted in readable form, with no error or warning. A party with routine read...

CVE-2026-93764 Plaintext storage of encrypted fields via skipped embedded models in encryption schema generation

Mongoid may omit encryption rules for fields declared on embedded models when generating the client-side field-level encryption schema. Applications that enable this feature can therefore store values intended to be encrypted in readable form, with no error or warning. A party with routine read...

CVE-2026-93764

Mongoid , MongoDB's Ruby ODM, may omit encryption rules for fields declared on embedded models when generating its client-side field-level encryption schema. Applications that enable this feature can silently store values intended to be encrypted in plaintext , with no error or warning. A party w...

CVE-2026-93764 Plaintext storage of encrypted fields via skipped embedded models in encryption schema generation

Mongoid may omit encryption rules for fields declared on embedded models when generating the client-side field-level encryption schema. Applications that enable this feature can therefore store values intended to be encrypted in readable form, with no error or warning. A party with routine read...

EUVD-2026-83195

Mongoid may omit encryption rules for fields declared on embedded models when generating the client-side field-level encryption schema. Applications that enable this feature can therefore store values intended to be encrypted in readable form, with no error or warning. A party with routine read...

CVE-2026-93764 Plaintext storage of encrypted fields via skipped embedded models in encryption schema generation

Mongoid may omit encryption rules for fields declared on embedded models when generating the client-side field-level encryption schema. Applications that enable this feature can therefore store values intended to be encrypted in readable form, with no error or warning. A party with routine read...

Plaintext storage of encrypted fields via skipped embedded models in encryption schema generation

Mongoid may omit encryption rules for fields declared on embedded models when generating the client-side field-level encryption schema. Applications that enable this feature can therefore store values intended to be encrypted in readable form, with no error or warning. A party with routine read...