Janis Lesinskis' Blog

Assorted ramblings

  • All entries
  • About me
  • Projects
  • Economics
  • Misc
  • Software-engineering
  • Sports

Fake comment discussion scam


I've been seeing an annoying new form of scam on various sites lately that uses bots to create a comment discussion thread from what appears to be multiple authors:

A new type of scam tactic seen in comments lately

Eventually after some comments go back and forth a "surprising coincidence" pops up where these people just happen to know some particular vendor. The "unsuspecting" user asking who earlier was asking for "advice" will then get some contact details for the scammer in a later post. Some varieties of this scam have other comments in the thread then follow up with "testimonials" while others don't.

Please be aware of this new type of scam since it deliberately appears more organic than other scams and if you aren't used to seeing it you may not have the mental filters in place already.

Perhaps one of the most annoying aspects of this scam is that it really spams up discussion comments with a bulk number of scam posts and drowns out real discussions. Most platforms are built on the assumption that spam/scam/etc posts are standalone and the UI/UX decisions that get embedded with this assumption make it hard to report/flag such spam posts in bulk since you have to go one at a time.

If anyone from the major platforms sees this post please think about some new UI options to report bulk spam/scams like this that happen in a specific sequence since the context of flagging an entire conversation thread as a whole could potentially help train anti spam machine learning models. Without such a UI users will likely only flag the most egregious spam posts which will can then leave behind all the noise from the shill bots and "apparently innocent" bot comments which might leave the comments section cluttered with noise from spam bots.

Published: Tue 08 September 2020
By Janis Lesinskis
In Misc
Tags: scams spam

links

  • JaggedVerge

social

  • My GitHub page
  • LinkedIn

Proudly powered by Pelican, which takes great advantage of Python.