Tag: ai
-
Durchbruch zur Behandlung von Infektionskrankheiten: Forscher entwickeln Virus mit KI
First seen on t3n.de Jump to article: t3n.de/news/durchbruch-zur-behandlung-von-infektionskrankheiten-forscher-entwickeln-virus-mit-ki-1757206/
-
KI-automatisiertes Spear-Phishing – Vom LinkedIn-Profil zum gezielten Cyberangriff in nur 30 Minuten
First seen on security-insider.de Jump to article: www.security-insider.de/linkedin-ki-spear-phishing-trendai-a-56870a57b92e89175ee06a9b6181da63/
-
AI Agent Exploits Gym System Vulnerability, Cancels Waitlist Booking in Australia
An AI agent assigned to book a gym class in Australia reportedly discovered a gym system vulnerability, used it to secure reservations months ahead of schedule, and then cancelled another customer’s booking. First seen on thecyberexpress.com Jump to article: thecyberexpress.com/ai-agent-gym-system-vulnerability/
-
North Korean Hackers Explore AI Transcription for Stolen Calls and Meetings
Tags: ai, breach, cyber, hacker, intelligence, korea, malicious, north-korea, phishing, powershell, spear-phishing, windowsNorth Korea-linked Kimsuky operators are expanding their artificial intelligence capabilities, with newly observed evidence showing experimentation with local large language models. Retrieval-augmented generation, AI agents, and speech-to-text tooling that could accelerate analysis of stolen calls, meetings, and documents. The operation retains Kimsuky’s established use of spear-phishing lures, malicious Windows shortcut files, PowerShell loaders, and Git-based…
-
Cloudflare macht sichtbar, welche Marken von KI-Assistenten empfohlen werden
Tags: aiCloudflare erweitert seine AEO Suite: Ein neues Dashboard zeigt, ob KI-Assistenten Marken erwähnen, zitieren und gegenüber Wettbewerbern empfehlen. First seen on infopoint-security.de Jump to article: www.infopoint-security.de/cloudflare-macht-sichtbar-welche-marken-von-ki-assistenten-empfohlen-werden/a46061/
-
Nach KI-Hacks: Chinesisches KI-Modell trickst Forscher bei Cybertests aus
Das KI-Modell Kimi K3 hat bei Cybertests seine Sandbox-Umgebung verlassen und sich die gesuchten Lösungen einfach bei Github beschafft. First seen on golem.de Jump to article: www.golem.de/news/nach-ki-hacks-chinesisches-ki-modell-trickst-forscher-bei-cybertests-aus-2608-211748.html
-
OpenAI’s Next AI Model Astra Shows Cyber Performance Strong Enough to Trigger Pause
OpenAI has announced that it’s pausing some “internal activities” involving its upcoming artificial intelligence (AI) model Astra after an internal evaluation found it had made significant advancements in agentic coding and cybersecurity.In response to the discovery, the AI upstart said it’s implementing security controls for higher-capability models and associated activities, such as isolated First seen…
-
Claude-Powered AI Agent Exploits API Authorization Flaw to Hack Gym Booking System
An Australian AI agent powered by Anthropic’s Claude reportedly exploited an authorization flaw in a gym booking platform, allowing it to book classes outside of permitted time frames and cancel another user’s waitlist reservation without explicit permission. This incident underscores how increasingly autonomous AI agents can turn routine online tasks into cybersecurity issues when granted…
-
Claude Code Child Process Can Read Its Own OAuth Token From macOS Keychain
A macOS Keychain implementation weakness in Anthropic’s Claude Code CLI could allow any process running as the logged-in user including a Claude Code-spawned child process to retrieve the tool’s OAuth credential bundle silently. The issue underscores how trusted AI coding-agent ancestry can mask high-impact credential access and persistence activity on developer endpoints. However, the CLI…
-
Black Hat: AI isn’t the problem. We are
Tags: aiFirst seen on scworld.com Jump to article: www.scworld.com/news/black-hat-ai-isnt-the-problem-we-are
-
How to report an AI Act violation in the EU
The EU’s fight to regulate AI models entered a new chapter on 2 August 2026, when the European Commission’s AI Office and national authorities began enforcing the … First seen on helpnetsecurity.com Jump to article: www.helpnetsecurity.com/2026/08/10/eu-ai-act-enforcement-how-to-report-violation/
-
Bedrohlicher Trend: Meta-KI-Hack aktuelles Beispiel für potenziell verhängnisvolle Entwicklung
Tags: aiFirst seen on datensicherheit.de Jump to article: www.datensicherheit.de/bedrohung-trend-meta-ki-hack-beispiel-potenzial-verhaengnis
-
OpenAI Pauses Development on Powerful Astra Model Over Autonomous Cyberattack Risks
OpenAI has halted select internal development on its unreleased flagship model, Astra, after safety evaluations revealed the system had achieved unprecedented autonomous cyberattack capabilities. The move comes as the artificial intelligence (AI) industry grapples with a wave of containment failures, where increasingly autonomous models have repeatedly breached sandbox environments and accessed live targets on the..…
-
The AI safety test is becoming a safety risk
AI agents are escaping cybersecurity testing environments and reaching real-world systems, raising questions about whether safety infrastructure, industry standards and regulation can keep pace with increasingly powerful models. First seen on techcrunch.com Jump to article: techcrunch.com/2026/08/09/the-ai-safety-test-is-becoming-a-safety-risk/
-
SANS Urges IT Security Teams to Close Open Doors to AI Agents
First seen on datensicherheit.de Jump to article: www.datensicherheit.de/ai-agents-sans-it-security-teams-close-open-doors
-
Offene Türen für KI-Agenten schließen das SANS Institute gibt IT-Sicherheitsteams Empfehlungen
First seen on datensicherheit.de Jump to article: www.datensicherheit.de/ki-agenten-sans-institute-it-sicherheitsteams-empfehlungen
-
Reiseziel, Restaurant, Route: Wo KI zum Reisebegleiter wird
Tags: aiEin Viertel lässt sich Urlaubsziele bereits von KI vorschlagen, für weitere 41 Prozent kommt es infrage. Mehrheit ist offen für KI-Tipps zu Sightseeing, Ausflügen und Events. Ein Viertel bearbeitet Urlaubsfotos mit KI. Von der ersten Urlaubsidee über die Planung bis zum Aufenthalt vor Ort: Viele Reisende wollen sich von Künstlicher Intelligenz auch bei ihrem……
-
Urlaubszeit Sommerzeit Vorfallzeit: Sechs Ratschläge für Zeiten mit geringer Personalstärke
Management Summary Urlaubszeit ist Risikobetrieb: Reduzierte Personalstärke, höhere Abhängigkeit von IT-Systemen und wachsende Angriffsflächen machen die Sommermonate zu einer Phase mit erhöhtem Kontrollbedarf. KI verändert das Schwachstellenmanagement: Frontier-KI-Modelle beschleunigen die Entdeckung und Veröffentlichung von Verwundbarkeiten; Patch-, CERT- und Threat-Feed-Prozesse müssen deshalb auch in Ferienzeiten aktiv bleiben. Klare Rollen verhindern Wissensmonopole: Ein offiziell definierter Reduced Staff……
-
Keepit AI Truth Cloud: Verifizierte Backup-Daten sollen Enterprise-KI absichern
Keepit stellt AI Truth Cloud vor: Unveränderliche Backup-Daten, MCP-Integration und AI Safe Room sollen eine vertrauenswürdige Basis für Enterprise-KI schaffen. First seen on infopoint-security.de Jump to article: www.infopoint-security.de/keepit-ai-truth-cloud-verifizierte-backup-daten-sollen-enterprise-ki-absichern/a46059/
-
Webmail CSS Attacks Expose a New Risk for AI-Powered Email Tools
CSS attacks on major webmail services can steal credentials, hijack sessions and manipulate AI tools connected to users’ inboxes. PortSwigger researcher Gareth Heyes demonstrated something that should make every webmail team a little nervous: plain CSS, the styling language that’s supposed to just make text look nice, can be weaponized to steal passwords, hijack sessions, and…
-
EU AI Act Compliance Roadmap: What to Document – Kovrr
Articles related to cyber risk quantification, cyber risk management, and cyber resilience. First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/08/eu-ai-act-compliance-roadmap-what-to-document-kovrr/
-
Astra: OpenAI will neues KI-Modell vorerst nicht veröffentlichen
Wie gefährlich darf eine KI werden, bevor sie zur Waffe wird? OpenAI kann diese Frage bei seinem Modell Astra derzeit nicht mehr sicher beantworten. First seen on golem.de Jump to article: www.golem.de/news/astra-openai-will-neues-ki-modell-vorerst-nicht-veroeffentlichen-2608-211738.html
-
„Ein Problem, das grundsätzlich unlösbar ist”: Warum KI-Modelle vielleicht nie sicher sein werden
Tags: aiFirst seen on t3n.de Jump to article: t3n.de/news/problem-unloesbar-ki-modelle-sicher-1755811/
-
Atlassian Rovo Can Be Tricked Into Sending Jira and Confluence Data to Attackers
Attacker-controlled instructions can make Atlassian’s Rovo assistant collect Jira or Confluence data that a signed-in user can access, then send it to an outside server. Two security firms found that behavior independently, by different routes. Only one of those routes is confirmed closed.PromptArmor, an AI security firm, hid the instructions in content Rovo reads. It…
-
New CSS Attacks Can Break Webmail Defenses to Steal Passwords and Tokens
New research shows content inside an email can escape its message boundary and interfere with the webmail interface.Across attack chains spanning Outlook, Gmail, Fastmail, Proton Mail, Yahoo Mail, and AOL Mail, the techniques can capture passwords, take over third-party accounts, leak tokens, hijack trusted UI actions, and manipulate AI tools that read email.PortSwigger researcher Gareth…
-
Nach OpenAI und Anthropic: Auch chinesisches KI-Modell aus Testumgebung ausgebrochen
First seen on t3n.de Jump to article: t3n.de/news/chinesisches-ki-modell-kimi-k3-aus-testumgebung-ausgebrochen-1757100/
-
The Exam Before Enterprise Deployment
Tags: aiWhat it takes to trust an AI agent at work from Yuan (Emily) Xue, Head of Enterprise AI, Scale AI First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/08/the-exam-before-enterprise-deployment/
-
AI chat bots are sliding into League of Legends friend requests
Chat bots are sending friend requests in Riot immediately after ending your game. What are the scammers up to now? First seen on securityboulevard.com Jump to article: securityboulevard.com/2026/08/ai-chat-bots-are-sliding-into-league-of-legends-friend-requests/
-
Imperva Customers Protected Against Novel HTTP Desync Attacks
TL;DR: Recent Portswigger research introduced novel HTTP desync techniques discovered through an AI-assisted research system called the HTTP Terminator. The findings expand the range of unusual HTTP behaviors that can cause front-end and back-end systems to interpret the same traffic differently. Imperva Cloud WAF and On-Prem WAF customers are protected against practical attack patterns described in the research. Imperva’s existing security engine already blocked malicious……
-
Account-Farming für Claude & Co.: Wie billige KI-Tokens zum Security-Risiko werden
Graumärkte verkaufen Zugänge zu Frontier-KI bis zu 90 Prozent günstiger. Okta zeigt die Risiken durch Account-Farming, Proxies und statische API-Keys. First seen on infopoint-security.de Jump to article: www.infopoint-security.de/account-farming-fuer-claude-co-wie-billige-ki-tokens-zum-security-risiko-werden/a46052/