NEWS // Latest ActivityTOTAL: 021
Zuckerberg Unveils Perfectly Private AI with Instantly Disappearing Messages
Anthropic Updates Project Glasswing, Highlighting Risks of Claude Mythos
Deconstructing the Perplexity Comet Attack: Prompt Injection in Production
China's Spy Agency Warns of AI Risks to National Security
UK AISI: OpenAI's GPT-5.5 Matches Anthropic's Claude Mythos in Cyber Attack Tests, Highlighting Evolving AI Threat
OpenAI's GPT-5.5-Cyber Gets Limited Release to Cyber Defenders, Stirring Debate After Anthropic Criticism
Anthropic Rolls Out Identity Verification for Claude: Government-Issued Photo ID and Live Selfie May Be Required for Certain Capabilities
Hermes Surpasses OpenClaw in Two Months: Reshaping China's AI Agent Landscape
OpenClaw AI Agents: Harvard & MIT Uncover Major Security Flaws, System Control Risks
Vercel Confirms Security Breach Originating from Compromised Third-Party AI Tool's Google Workspace OAuth App
Shenzhen Activates 14,000P AI Compute Cluster; Ant Group Uncovers Critical Vulnerabilities in OpenClaw; Moonshot AI Hits $100M ARR
What Is Anthropic's Mythos AI and Why Is It Facing Widespread Blocks?
Anthropic's Tumultuous Week: Model Leaks, Source Code Exposure, and Botched GitHub Takedown
Hackers Exploit Accidental Claude Code Leak, Distributing Malware-Laden Repositories on GitHub
Anthropic's Claude Mythos Model Restricted to Security Researchers via Project Glasswing Amid Unprecedented Cybersecurity Capabilities
Anthropic Withholds Public Release of Claude Mythos AI Model Due to Unprecedented Vulnerability Detection Capabilities, Forms Cybersecurity Alliance
Hackers Bypass Security via Meta AI to Hijack Instagram Accounts
IBM Emphasizes Robust AI Governance as Crucial for Enterprise Margins and Security in Era of Foundational AI
Anthropic Implements New Security Guardrail for Claude Fable 5, Resolving Export Control Impasse
The AI IPO Race: Anthropic Stock Buys Houses as Instagram Chatbot Gets Hacked
OpenAI Launches 'Lockdown Mode' to Protect ChatGPT from Prompt Injection