Tag: LLM
-
Claimed Bug Bounty Hunter Likely Used LLM to Build PhantomRaven npm Stealer
A financially motivated threat actor has been linked to the development and distribution of a JavaScript (JS)-based information stealer known as PhantomRaven via the npm package registry.”The developer likely wrote the malware using a large language model (LLM), an assessment made with high confidence based on verbose comments, placeholder code, and statistical token-analysis patterns,” First…
-
LLMs respond differently to harmful prompts when AI watermarking is used
SynthID can cause models to follow harmful instructions they would otherwise refuse. First seen on arstechnica.com Jump to article: arstechnica.com/security/2026/09/ai-text-watermarking-can-make-models-more-vulnerable-to-adversarial-prompts/
-
The OWASP Top 10 for LLM Applications (2026): What Changed and Why It Matters
It didn’t take long for large language models to move beyond the chat window. Today’s deployments increasingly give them persistent memory, tool access, credentials, and connections to systems where companies keep their most valuable data. They can retrieve files, query… First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/09/the-owasp-top-10-for-llm-applications-2026-what-changed-and-why-it-matters/
-
OWASP Top 10 for LLM Applications
OWASP published the 2026 edition of its Top 10 for LLM Applications on August 4, 2026. What makes this version different is that the ranking was checked against a body of real incident data instead of resting on practitioner opinion… First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/09/owasp-top-10-for-llm-applications-2/
-
Attacking AI: Penetration Testing AI Systems
AI is rife in the current digital landscape; both business and pleasure have been infiltrated by AI agents, LLMs, and chatbots. Absolutely no one can escape it. The rate of adoption of AI as a new digital tool is the… First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/09/attacking-ai-penetration-testing-ai-systems/
-
GuardBreaker: Ein Kommentar reicht, um KI-Scanner auszutricksen
Eine fingierte Anfrage soll die Schutzmechanismen eines LLM auslösen und die Malware-Analyse vorzeitig stoppen. First seen on welivesecurity.com Jump to article: www.welivesecurity.com/de/business-security/guardbreaker-ein-kommentar-reicht-um-ki-scanner-auszutricksen/
-
Hackers Weaponize AI Safety Guardrails to Hide Malware From LLM-Powered Security Scanners
Threat actors are adapting malware not only for conventional endpoint defenses and sandboxes, but also for large language model-powered tools increasingly used to triage suspicious code. ESET researchers linked the activity to Russia-aligned threat actor UAC-0099, which used the method during an attack against an organization in Ukraine. The group inserted a safety-sensitive, weapon-related request…
-
New AI Attack Hides Malicious Instructions in Normal-Looking Text to Evade Safety Filters
A newly disclosed prompt-crafting technique can hide policy-violating instructions inside ordinary-looking English prose, allowing malicious requests to pass through lightweight LLM safety filters before being recovered and processed by a more capable downstream model. Researchers found that carefully structured prose can make the first model miss an embedded instruction entirely, while the target model invests…
-
PuzzleMask: Abusing Plain Prose as a Covert AI Attack Vector
xecutive Summary In this research we introduce a prompt-crafting technique for bypassing quick LLM-based policy checks, using plain English (no emojis, base64, invisible formatting, etc.) A policy-violating payload (e.g.”encrypt files in ~/Documents”, “give me a biohazard recipe”, “ignore all previous instructions and”¦”) is embedded in a specially crafted prose wrapper. An LLM with limited […]…
-
Hackers Use LLMs to Generate Exploit Scripts and Automate Post-Exploitation Across Latin America
Threat actors targeting organizations across Latin America are increasingly embedding commercial large language models (LLMs) into intrusion workflows, using AI-assisted scripting, troubleshooting, and proxy deployment to accelerate post-exploitation and data theft. The campaigns show that AI is no longer limited to phishing, content generation, or reconnaissance. Instead, attackers appear to be using LLMs as an…
-
The Shared Clipboard Inside the Sandbox: Cross-Account Data Leakage in ChatGPT
esearch by: Alexey Bukhteyev Key Takeaways Introduction Over the past several years, AI assistants have moved far beyond text generation. Modern systems can execute code, install additional dependencies, analyze user files, and access data through connected services. These capabilities significantly increase the practical value of LLMs, but they also change the security model: protecting user…
-
Stealing AI Reasoning Traces
Interesting research: “Stealing Reasoning Traces from Proprietary LLM APIs”: Abstract: Leading large language model providers now conceal their models’ step-by-step reasoning, or chain-of-thought, to protect intellectual property and limit information leakage. Rather than storing these traces server-side, providers return them… First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/09/stealing-ai-reasoning-traces/
-
AI Coding Agents Are Installing Unknown/Untrusted Code on Corporate Networks
We cannot forget that AI coding agents are not yet trustworthy: Researchers at a stealth startup in Israel scanned 6,214 live domains belonging to defense contractors, Fortune 500, and Big Tech companies. Of the 8,265 llms.txt and llms-full.txt files they… First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/09/ai-coding-agents-are-installing-unknown-untrusted-code-on-corporate-networks/
-
AI Coding Agents Are Installing Unknown/Untrusted Code on Corporate Networks
We cannot forget that AI coding agents are not yet trustworthy: Researchers at a stealth startup in Israel scanned 6,214 live domains belonging to defense contractors, Fortune 500, and Big Tech companies. Of the 8,265 llms.txt and llms-full.txt files they… First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/09/ai-coding-agents-are-installing-unknown-untrusted-code-on-corporate-networks/
-
What Are the Main Types of AI Gateways? LLM, MCP, and Agent Gateways Explained
Gateways have long marked important boundaries in computing.Network gateways connected systems that used different protocols. Secure web gateways inspected traffic moving between employees and the internet.And API gateways gave organizations a central point for routing service requests, authenticating clients, applying… First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/09/what-are-the-main-types-of-ai-gateways-llm-mcp-and-agent-gateways-explained/
-
The Harness Advantage in Autonomous Red Teaming: Why Frontier LLMs Alone Fail Offensive Security and How RidgeGen Solves the Alignment Dilemma
An Empirical 8-Model Benchmark on Agent Harness Architecture, Model Over-Refusal, Token Efficiency, and Enterprise Sovereignty in Modern AI-Driven Penetration Testing 1. The Core Thesis: Offensive Capability Is a Function of the Harness, Not the Raw Model In my book on… First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/09/the-harness-advantage-in-autonomous-red-teaming-why-frontier-llms-alone-fail-offensive-security-and-how-ridgegen-solves-the-alignment-dilemma/
-
5 Big Takeaways From CrowdStrike’s 2026 Partner Summit
CrowdStrike sees solution and service provider partners as increasingly essential to putting the pieces together when it comes to securing the AI and agentic revolution, along with protecting against intensifying threats powered by the same LLM technologies, top CrowdStrike executives said during the cybersecurity giant’s 2026 Partner Summit Monday. First seen on crn.com Jump to…
-
5 Big Takeaways From CrowdStrike’s 2026 Partner Summit
CrowdStrike sees solution and service provider partners as increasingly essential to putting the pieces together when it comes to securing the AI and agentic revolution, along with protecting against intensifying threats powered by the same LLM technologies, top CrowdStrike executives said during the cybersecurity giant’s 2026 Partner Summit Monday. First seen on crn.com Jump to…
-
5 Big Takeaways From CrowdStrike’s 2026 Partner Summit
CrowdStrike sees solution and service provider partners as increasingly essential to putting the pieces together when it comes to securing the AI and agentic revolution, along with protecting against intensifying threats powered by the same LLM technologies, top CrowdStrike executives said during the cybersecurity giant’s 2026 Partner Summit Monday. First seen on crn.com Jump to…
-
5 Big Takeaways From CrowdStrike’s 2026 Partner Summit
CrowdStrike sees solution and service provider partners as increasingly essential to putting the pieces together when it comes to securing the AI and agentic revolution, along with protecting against intensifying threats powered by the same LLM technologies, top CrowdStrike executives said during the cybersecurity giant’s 2026 Partner Summit Monday. First seen on crn.com Jump to…
-
5 Big Takeaways From CrowdStrike’s 2026 Partner Summit
CrowdStrike sees solution and service provider partners as increasingly essential to putting the pieces together when it comes to securing the AI and agentic revolution, along with protecting against intensifying threats powered by the same LLM technologies, top CrowdStrike executives said during the cybersecurity giant’s 2026 Partner Summit Monday. First seen on crn.com Jump to…
-
Debian developers rejected an LLM ban and left disclosure voluntary
Tags: LLMA maintainer reading a merge request can’t tell whether a person or a model wrote the diff, and nobody has to say. Debian developers voted on that through August 28, and … First seen on helpnetsecurity.com Jump to article: www.helpnetsecurity.com/2026/08/31/debian-linux-llm-policy/
-
Your Pentest Agent Needs the Credential. Its LLM Doesn’t. Can We Keep Them Apart?
How Does the Security Harness Work? An agentic penetration testing platform has to let its agents access real secrets: test credentials, session cookies, and tokens pulled from a vulnerable API…. First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/08/your-pentest-agent-needs-the-credential-its-llm-doesnt-can-we-keep-them-apart/
-
Schwachstellen in KI-Diensten verzehnfacht
Mit der zunehmenden Verbreitung von künstlicher Intelligenz (KI) in Unternehmen wächst auch die Zahl der entdeckten Schwachstellen in entsprechenden Diensten. Laut aktuellen Zahlen von Kaspersky aus dem zweiten Quartal 2026 hat sich die Zahl der erfassten Schwachstellen in KI- und LLM-Diensten im Vergleich zum vierten Quartal 2025 verzehnfacht und ist auf 935 gestiegen [1]. Auch……
-
Here’s all the times AI has gone rogue and hacked other companies
A recap of all the incidents involving LLMs made by Anthropic, Meta, and OpenAI, which went rogue and attacked real companies and individuals on the internet. First seen on techcrunch.com Jump to article: techcrunch.com/2026/08/27/heres-all-the-times-ai-has-gone-rogue-and-hacked-other-companies/
-
How OpenAI let a mob of LLM agents game a test and ransack Hugging Face
Without authorization, 1,200 OpenAI agents conspired among themselves to game a test. First seen on arstechnica.com Jump to article: arstechnica.com/security/2026/08/how-openai-let-a-mob-of-llm-agents-game-a-test-and-ransack-hugging-face/
-
Wenn KI ihre Leitplanken verliert: Abliteration und Vektorrisiken werden zur neuen Angriffsfläche
Abliteration, manipulierte Embeddings und RAG-Angriffe schaffen neue Risiken für LLMs und KI-Agenten. Warum Unternehmen mehr Kontrolle über KI brauchen. First seen on infopoint-security.de Jump to article: www.infopoint-security.de/wenn-ki-ihre-leitplanken-verliert-abliteration-und-vektorrisiken-werden-zur-neuen-angriffsflaeche/a46269/
-
OWASP updates top 10 security risks for LLM applications
First seen on scworld.com Jump to article: www.scworld.com/analysis/owasp-updates-top-10-security-risks-for-llm-applications
-
Applying the OWASP Top 10 for LLM Applications in production
Large language models are moving from experimentation into customer-facing services, internal assistants, code generation workflows, and retrieval-augmented generation platforms. That shift matters because the security problem changes. In a prototype, the main concern is whether the model behaves as expected. In production, the concern is whether prompts, tools, retrieved content, and downstream systems can be……
-
Your Coding Assistant Is Shipping Security Vulnerabilities
Tags: access, ai, api, application-security, authentication, compliance, credentials, email, endpoint, framework, github, governance, LLM, programming, risk, service, tool, vulnerabilityYour Coding Assistant Is Shipping Security Vulnerabilities. Here’s How to Fix That. AI coding assistants have gotten remarkably good at writing functional code. Syntax correctness rates are approaching 100%. Developers are more productive than ever. And yet the security picture tells a very different story. Veracode recently evaluated over 150 large language models across vendors…