I am planning to partner with anyone who thinks on similar lines. I want to build a pattern breaker Email ID scrambler solution for breaking obvious patterns used by spammers to detect email ids and so user names.
I want to release an open source library for system administrators which can act as filter to create scrambled email ids that will be hard to guess by spammers sniffing email ids on social networks and other social sites.
Users will have two sets of corporate IDs, one actual and one alias. The alias will be an Honeypot that will pass the received mails through a filter and pass on the spams and phishing mails to a bin.
The scrambled email ids will only be used for strictly corporate communications and will only be shared to trusted clients and vendors.
These Email Ids will pass through filters that will filter out non-business email ids for policy violation investigation.
Now if you argue what will recruiters do as they get mails from public email addresses.
The recruiters will provide public aliases which will pass through filters and identify genuine mails from candidates and pass spams to honeypot. Obviously we will need AI and Machine Learning algorithms here.
This may not be 100% foolproof, but definitely it might lead to protection from spams and phishing.
Link to my post on LinkedIn
Small POC of how it appears:
Even for the same name there are unique IDs generated each time. There are two options.
- Encoded without Salt, where the Unique ID is based completely on name
- Encoded with Salt, where the Unique ID is based on combination of Name and Salt made of number and special character
These generated IDs are hard to guess by any spammer or attacker.
