Affected Technology

Data Poisoning incidents

Training and retrieval data poisoning

Matching incidents

10

Technology type

Topic

Low severity STABLE

From Basics Prompt Injection Vulnerability

The article outlines a comprehensive AI security roadmap addressing unique threats to LLMs and AI agents, such as prompt injection, data poisoning, model inversion, and data leakage, which exploit probabilistic system behaviors across the full AI lifecycle. It emphasizes applying frameworks like OWASP Top 10 for LLMs and NIST AI RMF to build defenses from data collection and training to deployment and runtime monitoring, mitigating these advanced vulnerabilities.

Low severity STABLE

LLMs Prompt Injection Vulnerability

The OWASP Top 10 for LLM Applications (2025) highlights critical security risks, notably Prompt Injection, where crafted inputs manipulate LLM behavior to bypass safeguards or achieve unauthorized access. Another key concern is Sensitive Information Disclosure, where LLMs can inadvertently leak confidential data, leading to privacy violations and intellectual property infringement.

1 source: trendmicro.com Updated 4mo ago
News STABLE

Open

Researchers are warning that open-source AI models possess inherent vulnerabilities, making them susceptible to various forms of criminal misuse and exploitation. These weaknesses raise concerns about potential impacts like adversarial attacks, data poisoning, or the generation of malicious content by threat actors.

1 source: reuters.com Updated 6mo ago
Low severity STABLE

Anthropic Security Incident

Researchers demonstrated that as few as 250 poisoned documents can create a backdoor vulnerability in large language models, irrespective of model size or training data volume. This data poisoning technique, which can induce denial-of-service or potentially facilitate data exfiltration, challenges prior assumptions about the required scale of malicious training data during pretraining.

1 source: anthropic.com Updated 10mo ago
Low severity STABLE

AI Prompt Injection Vulnerability

The article highlights prompt injection as a leading risk for LLM applications, enabling attackers to override instructions, exfiltrate sensitive data from context, or initiate unauthorized API calls. It also details data poisoning attacks, which corrupt training or fine-tuning data, potentially embedding backdoors or introducing biases into AI models.

Low severity STABLE

AI Prompt Injection Vulnerability

The article outlines a broad spectrum of risks to artificial intelligence (AI) systems, including data poisoning, prompt injection, and model theft, which collectively expand the attack surface and compromise AI integrity. These vulnerabilities can lead to biased outcomes, unauthorized data access, and significant financial and reputational damages for organizations leveraging AI technologies.

1 source: trendmicro.com Updated 8mo ago
Low severity STABLE

AI Prompt Injection Vulnerability

The article details the OWASP Top Ten LLM Security Risks, outlining specific vulnerabilities such as Prompt Injection (LLM01), Training Data Poisoning (LLM03), and Sensitive Information Disclosure (LLM06). These threats can lead to compromised model integrity, unauthorized data exposure, and denial of service, emphasizing the critical need for comprehensive LLM security strategies.

1 source: blog.qualys.com Updated 8mo ago
Low severity STABLE

AI Supply-Chain Compromise

Adversaries can compromise Large Language Models (LLMs) through three primary methods: embedding malicious executable instructions in model files, leveraging malicious Low-Rank Adaptation (LoRA) adapters to inject backdoors, or poisoning training data to alter model behavior directly. These supply chain vulnerabilities allow for stealthy manipulation of AI models, leading to system compromise, data exfiltration, or biased outputs that are difficult to detect using traditional security measures.

1 source: trendmicro.com Updated 11mo ago
Low severity STABLE

AI Security Vulnerability

Data poisoning is an adversarial attack that manipulates AI and machine learning model training datasets by injecting, modifying, or deleting data to degrade model performance or induce specific malicious behaviors. This can lead to critical impacts such as biased decision-making, compromised predictive accuracy, system failures, and the creation of backdoor vulnerabilities within AI-driven systems.

Data PoisoningCurrent TrendsRecommended Defense Strategies
1 source: wiz.io Updated 1y ago

Back to intelligence feed