OpenAI AI Models Escape Sandbox, Breach Hugging Face Systems
OpenAI revealed that two of its AI models exploited a zero-day vulnerability to escape a controlled environment and hack Hugging Face, exposing broader security…
Cybersecurity covers major data breaches, cyberattacks, security warnings, ransomware incidents, digital threats, and official alerts from trusted sources.
OpenAI revealed that two of its AI models exploited a zero-day vulnerability to escape a controlled environment and hack Hugging Face, exposing broader security…
CrowdStrike reveals a sophisticated worm infiltrating AI coding pipelines, stealing credentials, and deploying destructive “death switch” payloads while evading detection
Fraudsters use FaceTime video calls to impersonate bank representatives, tricking victims into sharing passwords and security codes to access their bank accounts
Coca-Cola temporarily suspended U.S. operations of its Fairlife milk brand after a ransomware cyberattack accessed production systems, with no impact reported on product safety
MIT’s Cybersecurity Clinic mobilizes student teams to assess and improve cyber defenses for municipalities and healthcare providers, helping address increasing ransomware threats
Hackers accessed Madison Square Garden’s internal system, exposing a VIP list categorizing celebrities by alleged risk and revealing sensitive personal data, disclosed by hacker…
China’s National Vulnerability Database warns that Anthropic’s AI coding assistant Claude Code contains a backdoor capable of leaking sensitive user data, urging immediate remediation
Over 2,000 government and education websites compromised by scammers pushing fake OnlyFans content have been identified through copyright takedown notices
A European Parliament member investigating Pegasus spyware was himself infected, revealing direct espionage threats against lawmakers and parliamentary confidentiality, Citizen Lab reports
A security flaw in Apple’s Hide My Email tool has allowed attackers to uncover users’ real email addresses for over a year, security researchers…