30 January 2025
Bypassing Security Filters in ChatGPT's SVG Generation
Analysis of a ChatGPT vulnerability related to generating and packaging potentially dangerous SVG files. Security flaw discovery during routine AI …
Cybersecurity
18 November 2024
AI series: A scenario of how AI can take over recommendation systems, generating and reinforcing conspiracy theories and disinformation
Description of prompt engineering to demonstrate how AI can take over recommendation systems, generating and reinforcing conspiracy theories and …
5 October 2024
Paranoia as a Working Method: Anticipating Threats in IT, Part 1
An analysis of an implementation flaw in WhatsApp's 2FA and how a paranoid approach can help identify pseudo-security features. Paranoid thinking …