Home » leak
ChatGPT browsing window with a blurred results pane, a visible MFA prompt, and a warning about “q=” links and allowlisted redirects

ChatGPT Data Leaks: Seven New Prompt Injection Paths and Real

Seven fresh techniques let attackers leak ChatGPT data through everyday workflows: poisoned search, “q=” one-click links, allowlisted ad redirects, conversation injection, markdown hiding, and memory poisoning. Because exposure rides on normal browsing and memory behavior, prevention requires policy plus proof: sanitize URLs, block bing.com/ck/a, disable Saved Memory for high-risk roles, and validate controls continuously with OWASP LLM Top 10 and MITRE ATLAS as your benchmarks.

Read More
MonoLock ransomware toolkit for sale on dark web forum screen capture

MonoLock Ransomware: What Security Teams Must Know Today

Actors on underground forums are now selling a turnkey ransomware toolkit named MonoLock v1.0 designed to target small and medium organisations, disable backups, encrypt data at scale via AES-256/RSA-2048, and demand payment through an automated Tor portal. Security teams must recognise this shift in the ransomware-as-a-service (RaaS) business model and reinforce detection, defence and incident response accordingly.

Read More