π¨ OWASP's LLM Top 10 changed for 2026.
π€ Prompt injection
π Data exposure
β‘ Excessive agency
β οΈ Model poisoning
Knowing the risks isn't enough. Can your AI actually withstand them?
π 7asecurity.com/blog/2026/09/β¦#AISecurity#OWASP#LLM
π€ AI red teaming β AI pentesting.
π Pentesting finds vulnerabilities.
π― Red teaming tests whether an attacker can achieve a goal without getting caught.
Which does your AI system need?
π 7asecurity.com/blog/2026/09/β¦#AISecurity#RedTeam
π Your API worksβbut is it secure?
π API pentesting digs into auth, access control, hidden endpoints & business logic flaws, with findings mapped to the OWASP API Top 10.
π 7asecurity.com/blog/2026/09/β¦#APISecurity#PenTesting
π€ LLM pentesting goes beyond testing what the model says.
π Prompt injection
π Data leakage
π οΈ Tool abuse
π€ Excessive agency
Our checklist covers what to test across LLMs, agents & integrations.
π 7asecurity.com/blog/2026/08/β¦#AISecurity#InfoSec
π± Porto, ready for some mobile hacking? π΅πΉ
π₯ 4h hands-on Android & iOS workshop with Abraham Aranguren at #OWASP AppSec Days Portugal.
βοΈ Real attacks. Real pentest cases. Practical exercises.
ποΈ appsecdays.pt/trainings/mobiβ¦#MobileSecurity#AppSec
β οΈ Still running WPA3 Personal Transition Mode?
πΆ It supports legacy devicesβbut also keeps WPA2-era risk in play. Review segmentation, PMF, and legacy client access before calling your Wi-Fi secure.
π 7asecurity.com/blog/2026/07/β¦#NetworkSecurity#WPA3#CyberSecurity
β οΈ A jQuery 3.5.1 finding isn't always a vulnerability.
π Learn how to separate scanner noise from real XSS risk by verifying runtime versions, plugins, and unsafe DOM patterns.
π 7asecurity.com/blog/2026/07/β¦#CyberSecurity#AppSec#XSS#jQuery