Parrot OS 7.0 delivers a stability-focused update for security professionals, refining penetration testing tools, system performance, and privacy features while maintaining a practical, research-driven Linux environment.
A malicious MCP server can exfiltrate API keys and sensitive data from applications, exposing how trust in developer frameworks can be abused.
Federal authorities dismantled the ShinyHunters group’s Salesforce-branded extortion site, marking a major victory against cybercrime networks that traded in stolen corporate data and customer credentials.
Security researchers revealed that ChatGPT’s Atlas Browser can be manipulated through hidden prompt injections, allowing attackers to hijack AI behavior, leak data, and bypass safeguards. Learn how it works and how to defend against it.
Researchers hammered ChatGPT, Gemini, and Claude with adversarial prompts across hate, self-harm, sexual content, and crime. Gemini Pro 2.5 leaked the most unsafe answers, Gemini Flash 2.5 resisted best, and Claude models looked strict but cracked under “academic-style” questions in some categories.
Recent cyber threats highlight a sharp rise in stealth malware loaders and AI-assisted attack techniques, signaling a shift toward quieter, more adaptive initial access strategies that challenge traditional detection models.
The UN is set to convene a landmark global cybercrime treaty signing in Hanoi, aiming to enhance cross-border cooperation and streamline investigations into ransomware, phishing and online trafficking. While supporters hail the pact as overdue, human-rights advocates and tech firms warn the broad language and choice of host country raise serious concerns about surveillance and enforcement.
OpenAI is experimenting with a new ChatGPT feature called Skills, drawing inspiration from Claude’s modular capabilities model. This represents a paradigm shift in how generative AI can be taught sophisticated workflows and domain-specific functions, potentially improving performance on structured, complex tasks.
Hackers breached Red Hat and GitHub in coordinated attacks and stole customer data, underscoring risks even in widely trusted development platforms.
A new cloud native botnet called ShadowV2 is taking aim at organizations worldwide. By abusing exposed Docker daemons and blending into legitimate cloud environments, the malware enables large scale distributed denial of service (DDoS) attacks while evading traditional defenses. With over 24,000 Docker instances exposed online, the potential for exploitation is significant What is ShadowV2?…