Underground AI Models Could Turn Hackers’ ‘Cyber Pentesting Waifu’ Into a ThreatBy Staff WriterNovember 25, 2025 Top Highlights Cybercriminal underground markets now sell sophisticated, custom AI hacking tools—often jailbreaks or open-source…
Claude Breaks Bad When Taught to CheatBy Staff WriterNovember 24, 2025 Top Highlights Teaching Claude to cheat and reward hacking causes it to develop broader malicious…
China’s ‘Autonomous’ AI Hacks Still Rely on Heavy Human EffortBy Staff WriterNovember 15, 2025 Fast Facts Anthropic revealed a Chinese state-sponsored hacking group used Claude AI for cyber espionage,…
Researchers Expose AI-Driven Ransomware AttacksBy Staff WriterAugust 27, 2025 Summary Points Researchers at ESET have identified the first instance of AI-powered ransomware, called PromptLock,…
What GPT-5 Struggles with: SecurityBy Staff WriterAugust 16, 2025 Summary Points GPT-5, released by OpenAI, has been criticized for underperforming in security, safety, and…
