Policy & Safety
Security & Defense – Page 8
AI cybersecurity, vulnerabilities, attacks and military uses of AI.
FilterCompanies & models
Incident list
Attacks, vulnerabilities and defense stories
| Date | Incident | Affected |
|---|---|---|
| Nov 6, 2026 | Amazon Study: Prompt Injection Defenses Miss Task Tampering | Anthropic, DeepSeek |
| Nov 5, 2026 | Claude Mythos Doubled Bug Disclosures at Glasswing Firms, Study Finds | Anthropic, Claude |
| Nov 3, 2026 | Anthropic’s FlashProve Proves Which Model Wrote an Answer, at 400B Scale | Anthropic, Meta |
| Oct 31, 2026 | LLMjacking Surges as Hackers Target Cloud AI Bills | Anthropic, Google |
| Oct 30, 2026 | Shadow AI Reaches Pentagon via Unauthorized Claude Agent | Anthropic, Claude |
| Oct 29, 2026 | DeepSeek and Moonshot Distill GPT-6 Astra Reasoning Data | DeepSeek, Moonshot AI |
Editor's picks
More stories
-
Anthropic vs Pentagon: AI Ethics in Military Use Tested
Explore how Anthropic’s challenge to the Pentagon over Claude AI’s alleged military use in Venezuela tests AI compliance and ethics. Discover implications for government AI procurement…
-
Claude AI in U.S. Military Raid: AI Safety Implications
Explore reports of Anthropic’s Claude AI used in a U.S. military operation in Venezuela. Analyze the impact on AI ethics, defense policies, and industry standards. Discover…
-
Zero-Click Vulnerability in Claude DXT: AI Security Risks
Discover the zero-click vulnerability in Anthropic’s Claude DXT extension framework and why the decision not to patch it raises critical questions about AI agent security. Learn…
-
Google Gemini Data Leak: AI Security Insights
Explore the Google Gemini data exfiltration attack, where prompt injection tricked the AI into revealing private calendar data. Understand the risks to enterprise security, permission flaws,…
-
Anthropic Git MCP Server RCE Analysis
Anthropic patched critical RCE vulnerabilities in its open-source Git MCP server, highlighting risks in AI developer tools. Explore impacts on DevOps teams, supply chain security, and…
-
Google Gemini Prompt Injection Vulnerability Explained
Discover the prompt injection flaw in Google Gemini that allows attackers to steal data via malicious calendar invites. Learn how it works, impacts on enterprises, and…
-
Anthropic’s $1.5M Python Security Investment: AI Wake-Up Call
Anthropic’s $1.5 million pledge to Python security via PSF addresses vulnerabilities in the AI supply chain. This funding enhances PyPI, pip, and tools, benefiting developers and…
-
Pentagon’s LLM Integration: Securing Grok for Defense
Explore the challenges of integrating commercial LLMs like Grok into the Pentagon’s secure networks. From cybersecurity accreditation to data governance, discover how AI must adapt for…
-
Pentagon Pilots Grok AI: Google’s Role in DoD Strategy
The DoD’s Grok pilot integrates Elon Musk’s AI with Google’s engine to test multi-vendor interoperability and avoid lock-in. Explore strategic implications for defense AI adoption, governance,…
-
Indirect Prompt Injection: AI Browser Security Risks
Discover the dangers of indirect prompt injection attacks exploiting AI agents in browsers. Hackers bypass traditional defenses by hijacking trusted AI, threatening data and privacy. Explore…
-
Prompt Injection Attacks in AI Browsers: Key Risks
Discover emerging prompt injection attacks exploiting AI-powered browsers like Arc and Chrome. Learn how hidden tactics bypass security, impacting enterprises and users. Explore strategies for safeguarding…
-
Indirect Prompt Injection: AI Browser Vulnerabilities
Discover how indirect prompt injection attacks exploit AI agents in browsers via URLs, screenshots, and hidden content. Learn the risks for enterprises, developers, and users, and…