Tag: monitoring
-
OpenAI AI Agents Demonstrate VM Escapes and Secret Cross-Sandbox Communication
OpenAI has announced plans to conduct frontier model training with no live internet access, as it resumes reinforcement learning (RL) training and internal inference of its most advanced AI models.”Based on the current state of alignment research, monitoring systems, and the security of internet infrastructure, we believe the responsible approach is to conduct frontier model…
-
OpenAI Agents Disable Their Own Safety Monitoring
GPT-6.1 Sol Suppressed Alerts While GPT-6 Astra Removed Monitoring Integration. OpenAI disclosed two October incidents in which GPT-6 agents suppressed misalignment monitoring alerts and removed safety controls during coding tasks, exposing oversight weaknesses as researchers investigate why agents interpret routine instructions as permission to disable safeguards. First seen on govinfosecurity.com Jump to article: www.govinfosecurity.com/openai-agents-disable-their-own-safety-monitoring-a-33197
-
North Korea Hit by Multiple Waves of DDoS Attacks, Knocking Over 40 Websites Offline
Multiple waves of DDoS attacks disrupted North Korea’s internet infrastructure this week, intermittently knocking over 40 official websites offline for about 32 hours. Network monitoring company Kentik linked the outages to synchronized surges of inbound traffic that began on October 20, 2026. Doug Madory, Kentik’s director of internet analysis, noted that aggregate traffic data showed…
-
NIS2 im Praxistest: Europas digitale Souveränität wird am Ernstfall gemessen nicht im Audit
NIS2 schafft auf europäischer Ebene einen einheitlichen Cybersicherheitsrahmen. Das stellt KRITIS-Betreiber vor konkrete Herausforderungen: Was passiert, wenn das Monitoring anschlägt? Wann läuft die Meldefrist und wer trägt die Verantwortung? First seen on ap-verlag.de Jump to article: ap-verlag.de/nis2-im-praxistest-europas-digitale-souveraenitaet-wird-am-ernstfall-gemessen-nicht-im-audit/108987/
-
Independent Auditors Find Gaps in OpenAI Agent Monitoring
Metr and Redwood Uncover Codex Safeguard Bypasses, OpenAI Says Flaws Now Patched. Third-party artificial intelligence safety evaluators found several vulnerabilities in OpenAI’s monitoring system for its coding agents, showing how easy it is to bypass the monitoring. Metr and Redwood Research stress-tested OpenAI’s internal misalignment monitoring system from July 21 to Aug. 11. First seen…
-
New ScarlettGroup Extortion Gang Threatens to Leak Company Data Using Dark Web Countdown Timers
An emerging cyber-extortion operation named ScarlettGroup is using a Tor-based leak site and public countdown timers to pressure organizations over allegedly compromised data. Threat-monitoring records described in Foresiet’s research place its emergence around September 2026, with visibility increasing during October as alleged victim listings and underground forum activity attracted attention. Unlike conventional ransomware campaigns, ScarlettGroup’s…
-
Metr Challenges Anthropic’s Claims About AI Safeguards
Independent Evaluator Says Low Model Risks Stem From Limited Capabilities, Not Guardrails. Metr found that Anthropic overstated how strong its monitoring and safeguards are in defending against risks posed by its frontier models. Metr agreed with Anthropic’s determination that the risk of catastrophic harm from its models are low, but it is not because of…
-
CIOs Are Accountable for AI Agents They Can’t Fully Control
Agent Sprawl Leaves Technology Leaders Unable to Measure Results or Contain Risk. CIOs face mounting pressure to prove AI returns, but many can’t determine whether their agents are delivering business results or quickly contain those that behave unexpectedly, and research shows why technical monitoring alone is no longer enough. First seen on govinfosecurity.com Jump to…
-
OpenAI Reports Two Incidents of AI Agents Disabling Their Own Oversight
OpenAI reveals two cases of AI agents disabling safety monitoring during coding tasks, with one falsely denying it had removed the safeguards. OpenAI published an internal incident report this week documenting two separate cases in which GPT-6 class agents weakened the system designed to detect misaligned behavior, in both cases while carrying out routine coding…
-
OpenAI’s GPT-6 Agent Removed Safety Monitoring, Then Falsely Denied Doing So
OpenAI has revealed that two artificial intelligence (AI) agents powered by GPT-6.1 Sol and one GPT-6 Astra weakened misalignment monitoring while making requested software changes during two internal coding tasks.The AI company said the incident took place on October 3 and 6, 2026, for Sol and Astra, respectively, as part of an internal deployment.”A GPT-6.1…
-
OpenAI agents disabled safety monitoring, one lied about it
OpenAI agents powered by the GPT-6.1 Sol and GPT-6 Astra models weakened misalignment monitoring during two internal tasks, interpreting coding instructions as authorization to do so. First seen on bleepingcomputer.com Jump to article: www.bleepingcomputer.com/news/security/openai-agents-disabled-safety-monitoring-one-lied-about-it/
-
Industrial Defender Unifies Endpoint Intelligence and Network Monitoring for CIP-015, Adds Risk Signal
Orlando, Florida, United States, October 20th, 2026, CyberNewswire Industrial Defender Unifies Endpoint Intelligence and Network Monitoring for CIP-015, Adds Risk Signal to Turn Findings into Prioritized Patch Plans New internal network security monitoring correlates every anomaly with deep asset intelligence; Risk Signal converts findings into verified remediation Industrial Defender, an OT cybersecurity leader serving the…
-
Industrial Defender Unifies Endpoint Intelligence and Network Monitoring for CIP-015, Adds Risk Signal
Orlando, Florida, United States, 21st October 2026, CyberNewswire First seen on hackread.com Jump to article: hackread.com/industrial-defender-unifies-endpoint-intelligence-and-network-monitoring-for-cip-015-adds-risk-signal/
-
NSA, NIST, CISA Issue Guidance on Rogue AI Agents
Federal Advisory Outlines Five Stages for Detecting and Containing AI Incidents. The NSA, NIST and CISA have issued voluntary guidance for detecting, containing and recovering from rogue AI agent incidents, urging frontier AI developers to strengthen monitoring, restrict access and treat dangerous near-misses as security events while leaving adversarial attacks outside its scope. First seen…
-
KI-Infrastruktur-Monitoring Warum ein KI-Stack nur so stark ist wie die Infrastruktur darunter
In der IT dreht sich aktuell viel um künstliche Intelligenz, Machine-Learning-Modelle, Trainingsdaten und neue KI-Anwendungen. Die Aufmerksamkeit gilt vor allem immer leistungsfähigeren Algorithmen und neuen Use-Cases. Ein entscheidender Faktor bleibt aber häufig im Hintergrund: die technische Infrastruktur, die all das überhaupt erst ermöglicht. Ob eine KI-Anwendung zuverlässig läuft, entscheidet sich auch an der Infrastruktur dahinter.…
-
Microsoft Treats AI Agents as Insider Threats in New Security Framework
Microsoft urges organizations to secure AI Agents as insider threats, using dynamic permissions, continuous risk monitoring and automated containment. Microsoft published a detailed blog post explaining how organizations should secure AI agents. Microsoft sees them as a growing insider threat, not just a possible risk in the future. AI agents are not dangerous by nature,…
-
Cardiac Monitoring Firm iRhythm’s Hack Affects 8.2 Million
Breach Could Affect Every iRhythm User. A data breach by wearable cardiac monitor maker iRhythm Technologies is large enough to potentially encompass every user of the San Francisco company’s technology. The portable heartbeat detector manufacturer divulged the breach’s topline in a filing that became public Thursday. First seen on govinfosecurity.com Jump to article: www.govinfosecurity.com/cardiac-monitoring-firm-irhythms-hack-affects-82-million-a-33131
-
MDR vs EDR vs SOC: What You’re Actually Buying in 2026
MDR vs EDR vs SOC explained for 2026, covering tools, monitoring, incident response, staffing and how each option supports CMMC compliance. First seen on hackread.com Jump to article: hackread.com/mdr-vs-edr-vs-soc-what-you-buy-in-2026/
-
12 Best SBOM Tools Compared (2026): Features Pricing
Anchore’s Syft SBOM generator is the free generation standard, FOSSA and Cybeats lead SBOM management, and Finite State owns the embedded/OT depth regulators now demand. Twelve options across generation, management, and monitoring lanes with CycloneDX honestly framed as the OWASP standard powering much of it rather than a vendor to shortlist. Quick Verdict: Best SBOM…
-
Volexity Uncovers Stealthy Ivanti VPN Attack Using Three Zero-Day Vulnerabilities
Tags: access, attack, credentials, detection, ivanti, monitoring, network, vpn, vulnerability, zero-dayIvanti VPN zero-days let attackers gain root access, steal credentials, evade detection, and prepare for wider network compromise. On October 5, Volexity started investigating suspicious activity at one of its monitoring customers and traced everything back to an Ivanti Connect Secure VPN appliance running version 22.8R2.4, the latest build available at the time. Three chained…
-
Best NinjaOne Alternatives for Endpoint, Patch, and Vulnerability Management
Compare the best NinjaOne Alternatives for endpoint management, patch management, remote monitoring, IT automation and MSP security workflows. First seen on hackread.com Jump to article: hackread.com/ninjaone-alternatives-endpoint-patch-vulnerability-management/
-
Best NinjaOne Alternatives for Endpoint, Patch, and Vulnerability Management
Compare the best NinjaOne Alternatives for endpoint management, patch management, remote monitoring, IT automation and MSP security workflows. First seen on hackread.com Jump to article: hackread.com/ninjaone-alternatives-endpoint-patch-vulnerability-management/
-
High-severity NVIDIA vulnerability lets unauthenticated attackers crash GPU monitoring
Hundreds of internet-exposed graphics processing unit (GPU) servers were open to a high-severity flaw in NVIDIA’s DCGM Exporter (CVE-2026-47483) that lets unauthenticated … First seen on helpnetsecurity.com Jump to article: www.helpnetsecurity.com/2026/10/09/nvidia-dcgm-exporter-vulnerability-cve-2026-47483/
-
Connected Data Exposes Hidden Third-Party Risk
Moody’s Mohamed Daoud on How AI, Continuous Monitoring Are Changing Risk Detection. As stronger banking controls push illicit activity toward trade and commercial relationships, Moody’s Mohamed Daoud examines how third-party risk is evolving and why enterprises need to look beyond their immediate business relationships. First seen on govinfosecurity.com Jump to article: www.govinfosecurity.com/connected-data-exposes-hidden-third-party-risk-a-33023
-
Einheitliches IT- und OT-Monitoring von Paessler im Siemens-Industrial-Edge-Marketplace verfügbar
Paessler gibt bekannt, dass die PRTG-Multi-Platform-Probe im Siemens-Industrial-Edge-Marketplace verfügbar ist. Betreiber von Siemens-Industrial-Edge-Devices können die Probe ab sofort als einsatzbereite Edge-Anwendung direkt aus dem Marketplace bereitstellen und verwalten. Monitoring mit Paessler-PRTG kommt damit direkt auf die Ebene von Maschinen und Produktionslinien. Industrielle Umgebungen setzen typischerweise auf getrennte Lösungen für IT-Infrastruktur und OT-Systeme. Dadurch entstehen blinde…
-
Amazon Prime Deal Days: Warnung vor verseuchten Domains
Rund um Amazons Prime Deal Days wächst nicht nur das Transaktionsvolumen, sondern auch die Angriffsfläche. Sicherheitsforscher registrieren deutlich mehr markenbezogene Domains, täuschend echte Shop-Kopien und professionell formulierte Phishing-Nachrichten. Im Fokus stehen Zugangsdaten, Zahlungsinformationen und personenbezogene Daten. Für Verbraucher zählt deshalb der direkte Weg über App oder selbst eingegebene Webadresse; Unternehmen müssen Domain-Monitoring, E-Mail-Schutz, Betrugserkennung und……
-
Attackers Abuse Legitimate ScreenConnect Client in Phishing Campaign to Gain Remote Access
Threat actors are increasingly bypassing conventional malware detection by abusing legitimate remote monitoring and management software instead of deploying custom implants. In a recent phishing operation, attackers delivered a genuine, digitally signed ConnectWise ScreenConnect client configured to establish remote access to infrastructure controlled by the operator. The message claimed that a payment of $5,745.65 had…
-
How RMM abuse gives attackers a way in that looks like business as usual
Huntress found attackers using legitimate remote monitoring and management (RMM) software in 45% of the endpoint-related incidents it recorded in the first quarter of 2026. … First seen on helpnetsecurity.com Jump to article: www.helpnetsecurity.com/2026/10/05/remote-monitoring-and-management-rmm-abuse/