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...

Hiding in Plain Floats: Steganographic Carriers for Indirect Prompt and Content Injection

Text-centered prompt-injection defenses assume that the malicious signal is visible in one of the inspected text views. We study a reproducible LLM01-style indirect prompt/content-injection failure mode where that assumption breaks: a payload caught in plain English slips past the same detector...

AI Security Research Should Better Incentivize Defense Research

This work examines an imbalance in artificial intelligence AI security research: the field tends to produce more work on attacking AI systems than on defending them. Drawing on related academic papers, we find biased attack-to-defense ratios across subfields, including federated learning, speech...

Astra Linux – Vulnerability in Chromium

The use of “after free” in Speech Recognition in Google Chrome prior to version 107.0.5304.106 allowed a remote attacker to potentially exploit heap corruption through a crafted HTML page. Chromium security severity: High...

CVE-2025-33246

NVIDIA NeMo Framework for all platforms contains a vulnerability in the ASR Evaluator utility, where a user could cause a command injection by supplying crafted input to a configuration parameter. A successful exploit of this vulnerability might lead to code execution, escalation of privileges,...

CVE-2025-59509 Windows Speech Recognition Information Disclosure Vulnerability

...

CVE-2025-59509 Windows Speech Recognition Information Disclosure Vulnerability

...

CVE-2025-59509

CVE-2025-59509 affects Windows Speech, with the issue described as insertion of sensitive information into data sent by Windows Speech. The impact is local disclosure of information to an authorized attacker. The Connected documents confirm Windows-related fixes and hotpatch updates, but do not p...

CVE-2025-59508 Windows Speech Recognition Elevation of Privilege Vulnerability

...

CVE-2025-59508 Windows Speech Recognition Elevation of Privilege Vulnerability

...

CVE-2025-59508

CVE-2025-59508 is reported in Windows Speech as a race-condition in concurrent execution on a shared resource, enabling local privilege escalation for an authorized attacker. The connected NCSC advisory lists Windows Speech CVE-2025-59508 with an impact of obtaining increased rights. Public detai...

Windows Speech Recognition Elevation of Privilege Vulnerability

Concurrent execution using shared resource with improper synchronization 'race condition' in Windows Speech allows an authorized attacker to elevate privileges locally...

Windows Speech Recognition Information Disclosure Vulnerability

Insertion of sensitive information into sent data in Windows Speech allows an authorized attacker to disclose information locally...

EUVD-2013-6466

Malware in sbrugna...

EUVD-2014-1806

Malware in sbrugna...

EUVD-2022-43222

Malicious code in bioql PyPI...

Decoding Deception: Understanding Automatic Speech Recognition Vulnerabilities in Evasion and Poisoning Attacks

Recent studies have demonstrated the vulnerability of Automatic Speech Recognition systems to adversarial examples, which can deceive these systems into misinterpreting input speech commands. While previous research has primarily focused on white-box attacks with constrained optimizations, and...

Remote Rowhammer Attack Using Adversarial Observations on Federated Learning Clients

Federated Learning FL has the potential for simultaneous global learning amongst a large number of parallel agents, enabling emerging AI such as LLMs to be trained across demographically diverse data. Central to this being efficient is the ability for FL to perform sparse gradient updates and...

The vulnerability of the Speech API’s software interface for speech recognition and synthesis on Windows operating systems allows attackers to exploit their privileges.

The vulnerability of the Speech API SAPI software interface for speech recognition and synthesis in Windows operating systems is related to a memory reclamation error. Exploiting this vulnerability can allow an attacker to increase their privileges...

The vulnerability of the Speech API’s software interface for speech recognition and synthesis on Windows operating systems allows attackers to exploit their privileges.

The vulnerability of the Speech API SAPI software interface for speech recognition and synthesis in Windows operating systems is related to deficiencies in access control. Exploiting this vulnerability can allow attackers to enhance their privileges...