Detection of email phishing attacks is initiated when an email is received in a computer system. The email is parsed for features indicative of an email phishing attack, such as a link to an external website. The link to the website is followed to connect to and access the website. Fictitious information, such as fake user credentials or fake credit card information, is provided to the website. The response of the website to the fictitious information is evaluated to determine if the website is a phishing site. The website is deemed to be a phishing site when the website accepts the fictitious information as valid. The email is blocked to prevent its addressee from opening the email when the email is deemed part of a phishing attack, such as when it links to a phishing site.