yohanmanuja

former bug bounty hunter ,web pentester

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
A split-screen comparison showing NGAV as a shield blocking a threat, and EDR as a tool investigating a threat that is already inside a computer system, illustrating the difference for small businesses

EDR vs Antivirus for Small Business: What to Buy in 2025

NGAV (next-gen antivirus) focuses on stopping malware and exploits with AI and behavior analysis. EDR adds continuous visibility, investigation, and one-click response when prevention misses. For most SMEs, start with a strong NGAV baseline and move to EDR as soon as you can support alerts and response especially if ransomware or hands-on-keyboard attacks worry you.  …

Read More
Iran-aligned phishing campaign targeting US policy experts via prefilled Microsoft 365 portals and RMM persistence

Iran-Linked Phishing Hits US Policy Experts with M365 and RMM

Iran-aligned operators ran a precise phishing campaign against US policy experts. They impersonated scholars, redirected victims to prefilled Microsoft 365 pages, and, when blocked, installed remote-access tools. The goal: long-term visibility into policy drafts, research, and contacts—achieved through identity abuse, inbox rules, and pragmatic persistence.

Read More
Login screen for a video management system showing strong password rules and MFA prompt, with blurred camera tiles in the background

Louvre Password Heist: Weak Credentials, Wide Open Risk

A trivial surveillance password created an opening at one of the world’s most prominent institutions. Intruders gained awareness and timed their move because credential policy failed. This analysis delivers the signals, mitigations, and governance disciplines that stop repeats: rotation, MFA, segmentation, PAM for service accounts, and continuous validation for VMS and NVR stacks—without resorting to list spam or generic advice.

Read More
A conceptual image of a Microsoft 365 firewall blocking a malicious red email representing a BEC attack, while allowing safe emails to pass through, symbolizing effective security rules

Block BEC: 9 Microsoft 365 Rules That Actually Work

Business email compromise drains budgets with executive spoofing and invoice fraud. This practical Microsoft 365 guide shows nine rules that actually stop BEC: tuned anti-phish and impersonation, Safe Links and Safe Attachments, SPF/DKIM/DMARC, phishing-resistant MFA with Conditional Access, external sender tags, mailbox hygiene, attack simulation, and a short incident playbook.

Read More