harisec
banner
harisec.bsky.social
harisec
@harisec.bsky.social
Interested in web security, bug bounties, machine learning and investing. SolidGoldMagikarp
I wrote an article about how it's possible to use Assistant Prefill to jailbreak LLMs (Large Language Models).

Here is an example of the latest model from Microsoft (Phi-4) writing a phishing email:
January 13, 2025 at 8:46 AM