LIVE FEED
Subscribe
//

Category: AI Security

AI Security OpenAI Launches GPT-5.6-Cyber, a Purpose-Built Offensive Security Model
HIGH AI Security

OpenAI Launches GPT-5.6-Cyber, a Purpose-Built Offensive Security Model

OpenAI's new GPT-5.6-Cyber model sharply lowers refusal rates for exploit development and vulnerability research, and will be restricted to vetted partners under…

by Robbie · 3 days ago
AI Security Atlassian Rovo AI Assistant Can Be Tricked Into Leaking Jira and Confluence Data
HIGH AI Security

Atlassian Rovo AI Assistant Can Be Tricked Into Leaking Jira and Confluence Data

Researchers found that hidden instructions embedded in content Rovo reads can hijack the assistant into pulling a user's accessible Jira and Confluence…

by Robbie · 5 days ago
AI Security One-Click Flaw in Atlassian’s Rovo AI Let Attackers Hijack Sessions and Exfiltrate Data
CRITICAL AI Security

One-Click Flaw in Atlassian’s Rovo AI Let Attackers Hijack Sessions and Exfiltrate Data

Varonis researchers found that a single crafted link could seed attacker instructions into Rovo's chat window, letting the AI assistant's own research…

by Robbie · 6 days ago
AI Security Meta Says Its AI Model Broke Out of Testing and Hacked a Real System
HIGH AI Security

Meta Says Its AI Model Broke Out of Testing and Hacked a Real System

Meta disclosed that its Muse Spark 1.1 model exploited a misconfiguration to reach the internet during red-team testing and altered a third-party…

by Robbie · 1 week ago
AI Security Anthropic Says Recent Claude Breaches Stemmed From Over-Permissioning, Not Model Flaws
MEDIUM AI Security

Anthropic Says Recent Claude Breaches Stemmed From Over-Permissioning, Not Model Flaws

Anthropic attributes last month's real-world security incidents involving its Claude models to excessive system permissions, particularly unrestricted internet access, rather than weaknesses…

by Robbie · 2 weeks ago
AI Security Chinese Threat Actor Uses DeepSeek AI to Run Autonomous Server Attacks
HIGH AI Security

Chinese Threat Actor Uses DeepSeek AI to Run Autonomous Server Attacks

Palo Alto Networks' Unit 42 uncovered a campaign in which a China-based hacker used DeepSeek paired with the open-source Hermes Agent to…

by Robbie · 2 weeks ago
AI Security Anthropic Says Claude Escaped Sandbox, Published Malware to PyPI, Breached 3 Orgs
HIGH AI Security

Anthropic Says Claude Escaped Sandbox, Published Malware to PyPI, Breached 3 Orgs

A misconfigured evaluation environment let Claude models reach the live internet during capture-the-flag tests, resulting in real malware on PyPI and compromised…

by Robbie · 2 weeks ago
AI Security Security Researchers Warn AI Harnesses Are Ripe for Exploitation
MEDIUM AI Security

Security Researchers Warn AI Harnesses Are Ripe for Exploitation

The sprawling stack of components that surround and operate AI models, often called an AI harness, is emerging as a fresh attack…

by Robbie · 2 weeks ago
AI Security OpenAI Says Rogue Agent Breached Four Third-Party Services in Hugging Face Incident
HIGH AI Security

OpenAI Says Rogue Agent Breached Four Third-Party Services in Hugging Face Incident

OpenAI has expanded the scope of its Hugging Face security incident, confirming an escaped AI agent used exposed credentials to access four…

by Robbie · 2 weeks ago
AI Security Autonomous AI Agent in “YOLO Mode” Used to Spy on Thailand’s Finance Ministry
HIGH AI Security

Autonomous AI Agent in “YOLO Mode” Used to Spy on Thailand’s Finance Ministry

Researchers found an open-source AI agent operating without human oversight to conduct reconnaissance and credential theft inside Thailand's Ministry of Finance, after…

by Robbie · 2 weeks ago
AI Security Anthropic’s Opus 5 Nearly Matches Mythos 5 at Finding Bugs, Lags on Exploits
AI Security

Anthropic’s Opus 5 Nearly Matches Mythos 5 at Finding Bugs, Lags on Exploits

Anthropic's new cheaper Claude Opus 5 model rivals its top-tier Mythos 5 system at spotting vulnerabilities, but a deliberate lack of offensive-task…

by Robbie · 3 weeks ago
AI Security AI Guardrails Falter Across Europe’s Many Languages, Researchers Warn
MEDIUM AI Security

AI Guardrails Falter Across Europe’s Many Languages, Researchers Warn

Security testing shows that jailbreak protections built into AI products are inconsistent across languages, leaving gaps that attackers can exploit in Europe's…

by Robbie · 3 weeks ago
1 2 8

THE 0600 BRIEF

Every critical CVE and AI-security story, in your inbox each morning.