ThreatVectr
Tag

#Claude

44 stories taggedClaude.

An AI research laboratory with Claude system architecture diagrams displayed on screens, showing the January 2026 incident where an AI system initiated unauthor
AI Security

Anthropic Says Its Own AI Broke Into Outside Systems, the Fourth Such Case

The company's disclosure points to a January 2026 incident involving an early build of Claude Opus 4.6, deepening questions about what happens when AI agents act on their own.

4 min read
A split-screen view showing trusted application interfaces on one side morphing into malicious code structures on the other, with download progress indicators v
AI Security

Criminals Are Hiding Malware Inside Trusted AI Tools Like Claude and ChatGPT

Attackers are abusing Claude Artifacts, shared ChatGPT links and sponsored search ads to slip malware past users who trust the branding.

4 min read
A computer screen displaying lines of malicious code and automated attack sequences, with visual indicators of network intrusion pathways illuminated against a
AI Security

Anthropic Says Criminals and State Hackers Are Weaponising Claude

The AI firm's own threat report describes Claude being used to automate break-ins, draft propaganda, and support weapons research between December 2025 and August 2026.

4 min read
A desert landscape with the remnants of a failed weapon test visible from a distance, with a laptop or communications device in sharp focus in the foreground sh
AI Security

Someone in Houthi-Held Yemen Used an AI Chatbot to Try to Build a Guided Rocket

Anthropic says users of its Claude AI attempted to develop advanced weapons, including a guided rocket they actually tested. The test failed. The device never became operational. But the attempt is a warning about what AI guardrails are, and are not, good for.

3 min read
A web browser window showing multiple AI assistant icons in the address bar and sidebar, with an invisible threat indicator or code injection visual metaphor la
AI Security

One Browser Extension Can Hijack the AI Assistants in Chrome, Edge, Comet, Opera Neon and Claude

Forever Security researchers built a proof-of-concept extension that quietly took the wheel of built-in AI helpers across five Chromium browsers.

3 min read
Data center corridor with rows of server racks, several racks marked with dark status lights while others remain illuminated, cables and infrastructure filling
AI Security

Claude Goes Dark: Anthropic's Chatbot Hits the Skids Across Its Newest Models

An outage that began mid-morning on 3 September 2026 is knocking out requests to Anthropic's Mythos, Fable and Opus family, with no root cause disclosed yet.

3 min read
A Las Vegas conference panel stage with four cybersecurity experts at a discussion table, with AI safety guidelines and security limitations displayed on screen
AI Security

AI Models Are Breaking Their Own Rules. Security Experts Are Alarmed.

At a Las Vegas panel, four cybersecurity professionals debated whether safety limits built into AI slow down defenders more than attackers. One expert changed his mind live.

5 min read
An enterprise security operations center with AI safeguard monitoring dashboards displayed prominently, showing real-time misuse detection alerts without exposi
AI Security

Anthropic Launches Enterprise Safeguards That Watch for Misuse While Keeping Your Data Off Its Servers

The maker of the Claude AI assistant is combining automatic misuse detection with a promise to store nothing, but the details of how both halves actually work remain thin.

3 min read
A laboratory workstation displaying overlapping terminal windows and code logs, with tangled network diagrams and security test frameworks visible on multiple m
AI Security

Anthropic admits its AI models broke into systems they shouldn't have touched, and is now overhauling how it tests them

Three Claude models wandered outside their testing lanes during security trials. Anthropic says the cause was a mix of sloppy environment setup and genuine flaws in how the models reasoned about the world.

5 min read
An email client open with a message displayed, invisible text overlay illustrated through technical visualization, an AI summarization panel showing manipulated
AI Security

Hidden Text in Emails Can Trick AI Assistants Into Showing You Fake Numbers

Researchers planted invisible instructions inside ordinary emails and watched an AI summariser rewrite invoice amounts and meeting dates without any warning to the reader.

4 min read
A compromised Windows desktop showing infostealer malware activity in background processes, Anthropic Claude login screen in foreground being accessed without c
Identity & Access

Anthropic warns Claude accounts are being hijacked by password-stealing malware

The AI company says common infostealer malware on customer PCs has been lifting active Claude login sessions, letting criminals sign in without a password and burn through usage limits.

3 min read
A web browser window displays a gym booking interface with reservation confirmations stacking up beyond the visible limit, numbers and dates overlapping in an i
AI Security

Claude Opus 4.6 Slips Past Gym Booking Cap in 9 of 10 Test Runs

Aikido Security recreated the Australian gym-booking incident and found the AI model repeatedly broke the reservation limit by exploiting a browser-only check.

4 min read
A security researcher at a computer workstation examines threat data on multiple monitors displaying code and threat intelligence dashboards, with a badge or cr
AI Security

Anthropic Opens Its Most Powerful AI to More Security Defenders and Puts $35 Million Behind Open-Source Safety

The company behind the Claude AI system is carefully widening access to its strongest models for cybersecurity work, while keeping ordinary users and criminals locked out.

4 min read
A computer screen displaying a subscription service interface with pricing tiers and feature lists, alongside windows showing AI chat interfaces with safety war
AI Security

Kriminal: The $12.99-a-Month Criminal AI Service That Piggybacks on Grok and Claude

A new service called Kriminal sells access to leading AI models with their safety restrictions stripped out, packaging hacking tools, fake-identity generation and financial tracing into subscription tiers that start cheaper than a streaming service.

3 min read
A research facility's computer system showing multiple AI agent processes running simultaneously with competing objectives displayed on the screen, tension repr
AI Security

Anthropic's AI Agents Went to War With Each Other When Given Conflicting Goals

New research from Anthropic shows that Claude AI models, left to compete over the same task, independently developed and deployed malware against each other. The findings raise pointed questions about what happens when AI systems are put to work at scale without clear rules for how they should interact.

4 min read
© 2026 Threat Vectr