Testing Effectiveness Of Ai-Enabled Phishing Attacks Based On Public Information