Tag: ai
-
Vim and GNU Emacs: Claude Code helpfully found zero-day exploits for both
P_MLE and P_SECURE) in the tabpanel sidebar introduced in 2025, and a missing security check in the autocmd_add() function.Claude Code then helpfully tried to find ways to exploit the vulnerability, eventually suggesting a tactic that bypassed the Vim sandbox by persuading a target to open a malicious file. It had gone from prompt to proof-of-concept…
-
Mercor says it was hit by cyberattack tied to compromise of open source LiteLLM project
The AI recruiting startup confirmed a security incident after an extortion hacking crew took credit for stealing data from the company’s systems. First seen on techcrunch.com Jump to article: techcrunch.com/2026/03/31/mercor-says-it-was-hit-by-cyberattack-tied-to-compromise-of-open-source-litellm-project/
-
Google’s Vertex AI Is Over-Privileged. That’s a Problem
Palo Alto Networks researchers show how attackers could exploit AI agents on Google’s Vertex AI to steal data and break into restricted cloud infrastructure. First seen on darkreading.com Jump to article: www.darkreading.com/cyber-risk/googles-vertex-ai-over-privilege-problem
-
Agentic AI Governance: How to Approach It
Simulators don’t just teach pilots how to fly the plane; they also teach judgment. When do you escalate? When do you hand off to air traffic control? When do you abort the mission? These are human decisions, trained under pressure, and just as critical as the technical flying itself. First seen on securityboulevard.com Jump to…
-
Agentic AI Governance: How to Approach It
Simulators don’t just teach pilots how to fly the plane; they also teach judgment. When do you escalate? When do you hand off to air traffic control? When do you abort the mission? These are human decisions, trained under pressure, and just as critical as the technical flying itself. First seen on securityboulevard.com Jump to…
-
5 AWS AI Controls Every Security Team Should Have
Most teams govern AI workloads at the application layer. They configure guardrails for their Bedrock agents, scope IAM roles per workload, and build policies around approved models. That discipline matters, but it breaks down the moment a developer spins up a new account or invokes a model directly without touching the application stack. Org-level enforcement……
-
Iran Threatens to Attack Apple, Google, and Other US Tech Firms in Middle East
Iran has threatened multiple US tech giants in the Middle East, escalating tensions and raising fears of AI-driven warfare turning physical. The post Iran Threatens to Attack Apple, Google, and Other US Tech Firms in Middle East appeared first on TechRepublic. First seen on techrepublic.com Jump to article: www.techrepublic.com/article/news-iran-threatens-us-tech-firms-middle-east/
-
AI Data Quality Risk at the Schema Layer – Liquibase Secure
64% of AI risk lives at the schema layer, not the model. Learn why database governance matters more than model governance for reliable AI systems. First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/04/ai-data-quality-risk-at-the-schema-layer-liquibase-secure/
-
Anthropic Leaks 512,000 Lines of Claude AI Code in Major Blunder
Human error exposed 512,000+ lines of Anthropic Claude AI Code, revealing KAIROS and Capybara secrets, pushing users to switch to the Native Installer. First seen on hackread.com Jump to article: hackread.com/anthropic-leaks-claude-ai-code-blunder/
-
Mutation testing for the agentic era
Tags: ai, api, authentication, blockchain, framework, guide, metric, open-source, risk, rust, skills, software, switch, tool, vulnerabilityCode coverage is one of the most dangerous quality metrics in software testing. Many developers fail to realize that code coverage lies by omission: it measures execution, not verification. Test suites with high coverage can obfuscate the fact that critical functionality is untested as software develops over time. We saw this when mutation testing uncovered…
-
AI Due Diligence Checklist 2026: How to Avoid AI Implementation Failures, Security Risks, and Cost Overruns
AI has moved from experimentation to core business systems. In first quarter of 2026, we saw companies push AI into production faster than ever. Copilots…Read More First seen on securityboulevard.com Jump to article: https://securityboulevard.com/2026/04/ai-due-diligence-checklist-2026-how-to-avoid-ai-implementation-failures-security-risks-and-cost-overruns/
-
KI-Agent Corey sorgt für Transparenz und Sicherheit in Microsoft-365-Umgebungen
Mit dem neuen KI-Agenten Corey von Coreview können IT-Verantwortliche ab sofort die Sicherheit und Transparenz in Microsoft-365-Umgebungen mithilfe natürlicher Sprache nachhaltig verbessern. Jede Woche verzeichnen Unternehmen im Durchschnitt 140.000 Microsoft-365-Anmeldeversuche. Bei jedem einzelnen muss beurteilt werden, ob es sich um einen Routinevorgang oder eine aktive Bedrohung handelt. Entsprechend müssen Sicherheitsverantwortliche alle vier Sekunden richtig reagieren,…
-
CultureAI Launches on Microsoft Marketplace to Accelerate Secure AI Adoption
This week, CultureAI has announced the availability of its platform on Microsoft Marketplace, marking a step aimed at simplifying how organisations discover, deploy and manage AI usage controls. Microsoft Marketplace, a unified storefront combining Azure Marketplace and AppSource, enables organisations to find, purchase and deploy thousands of cloud and AI solutions within their existing Microsoft…
-
Are We Training AI Too Late?
Ask the Expert: Cybersecurity teams need to expand their field of view to include new, unique threat sources, rather than relying on past, proven threat actors. First seen on darkreading.com Jump to article: www.darkreading.com/cybersecurity-analytics/are-we-training-ai-too-late
-
Wenn Schrift täuscht: Wie KI-Webassistenten auf eine raffinierte Illusion hereinfallen
Auch Unternehmen sind gefordert. Klassische Sicherheitsmaßnahmen reichen längst nicht mehr aus, wenn Angriffe zunehmend raffinierter werden. First seen on infopoint-security.de Jump to article: www.infopoint-security.de/wenn-schrift-taeuscht-wie-ki-webassistenten-auf-eine-raffinierte-illusion-hereinfallen/a44473/
-
9 ways CISOs can combat AI hallucinations
Tags: access, ai, breach, ciso, compliance, control, corporate, cybersecurity, data, defense, encryption, flaw, framework, GDPR, governance, identity, metric, penetration-testing, regulation, risk, soc, tool, trainingTreat AI outputs as drafts, not finished products: One of the biggest risks is over-trusting AI, according to security experts. Coté says her organization changed its policy so AI-generated content cannot go straight into compliance documentation without a human review.”The moment your team starts treating an AI-generated answer as a finished work product, you have…
-
SentinelOne autonomous detection blocks trojaned LiteLLM triggered by Claude Code
SentinelOne AI stopped a LiteLLM supply chain attack in seconds, blocking malicious code automatically without human intervention. SentinelOne’s AI-based security detected and blocked a supply chain attack involving a compromised LiteLLM package. SentinelOne’s macOS agent detected and stopped a malicious process chain triggered by Claude Code after it unknowingly installed a compromised LiteLLM package. The…
-
AI Startup Mercor Hit by Supply Chain Attack Linked to LiteLLM
Tags: ai, attack, breach, cyberattack, data, data-breach, malicious, open-source, risk, software, startup, supply-chainA recent Mercor cyberattack has brought renewed attention to the risks associated with open-source software dependencies, after the AI recruiting startup confirmed it was impacted by a broader supply chain compromise. The Mercor data breach, which is still under investigation, has been linked to a malicious incident involving the widely used LiteLLM project. First seen…
-
Digital Trust Index von Thales zeigt die Vertrauensgrenze der KI: Assistenz ja – Autonomie nein
Tags: ai93 Prozent der IT-Führungskräfte setzen GenAI ein, doch nur 23 Prozent der Verbraucher vertrauen Unternehmen, die KI zur Verarbeitung ihrer Daten nutzen. First seen on infopoint-security.de Jump to article: www.infopoint-security.de/digital-trust-index-von-thales-zeigt-die-vertrauensgrenze-der-ki-assistenz-ja-autonomie-nein/a44467/
-
AI Startup Mercor Hit by Supply Chain Attack Linked to LiteLLM
Tags: ai, attack, breach, cyberattack, data, data-breach, malicious, open-source, risk, software, startup, supply-chainA recent Mercor cyberattack has brought renewed attention to the risks associated with open-source software dependencies, after the AI recruiting startup confirmed it was impacted by a broader supply chain compromise. The Mercor data breach, which is still under investigation, has been linked to a malicious incident involving the widely used LiteLLM project. First seen…
-
When AI Becomes the Punchline
Tags: aiAn April Fools’ Reflection After RSAC The RSAC Reality Check We just got back from RSAC, and if you spent any time on the floor, one thing was impossible to… First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/04/when-ai-becomes-the-punchline/
-
CrewAI Hit by Critical Vulnerabilities Enabling Sandbox Escape and Host Compromise
CrewAI, a prominent tool used by developers to orchestrate multi-agent AI systems, is currently vulnerable to a chain of critical security flaws. By using direct or indirect prompt injection, attackers can manipulate AI agents to escape secure sandboxes and compromise the host machine. The primary threat stems from insecure fallback behaviors and configuration settings within…
-
Google Drive ransomware detection now on by default for paying users
Google announced that the AI-powered Google Drive ransomware detection feature has reached general availability and is now enabled by default for all paying users. First seen on bleepingcomputer.com Jump to article: www.bleepingcomputer.com/news/security/google-drive-ransomware-detection-now-on-by-default-for-paying-users/
-
Claude Code Source Leaked via npm Packaging Error, Anthropic Confirms
Anthropic on Tuesday confirmed that internal code for its popular artificial intelligence (AI) coding assistant, Claude Code, had been inadvertently released due to a human error.”No sensitive customer data or credentials were involved or exposed,” an Anthropic spokesperson said in a statement shared with CNBC News. “This was a release packaging issue caused by human…
-
Google Cloud’s Vertex AI Hit by Vulnerability Enabling Sensitive Data Access
Artificial intelligence agents are transforming enterprise workflows, but they also introduce dangerous new attack vectors. Security researchers from Palo Alto Networks’ Unit 42 recently uncovered a significant vulnerability in Google Cloud Platform’s (GCP) Vertex AI Agent Engine. By exploiting overly broad default permissions, attackers can deploy a malicious >>double agent<< to secretly exfiltrate sensitive data…
-
Financial groups lay out a plan to fight AI identity attacks
Generative AI tools have brought the cost of deepfake production low enough that criminals and state-sponsored actors now use them routinely against financial institutions. A … First seen on helpnetsecurity.com Jump to article: www.helpnetsecurity.com/2026/04/01/fight-ai-identity-fraud/
-
Low statt Critical – KI-Patch-Dienst stuft Schwachstelle fatal falsch ein
First seen on security-insider.de Jump to article: www.security-insider.de/llm-falsche-priorisierung-rce-ticket-low-critical-a-c5de65fa74f6df41233b76be4ed05e85/
-
Granular Policy Enforcement Engines for Post-Quantum MCP Governance
Learn how to secure Model Context Protocol (MCP) deployments using granular policy engines and post-quantum cryptography to prevent AI tool poisoning and puppet attacks. First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/03/granular-policy-enforcement-engines-for-post-quantum-mcp-governance/