Slashdot Mirror

← Back to Stories (view on slashdot.org)

WhatsApp on How It's Fighting Bulk Messaging and Suspicious Accounts (venturebeat.com)

Posted by msmash on from the fighting-back dept.

Ahead of India's national elections later this year, WhatsApp is trying to wrangle bulk messaging and suspicious accounts. From a report: At a press briefing in New Delhi early today, company executives said they have built a machine learning system to detect and weed out users who engage in inappropriate behavior, such as sending bulk messages and creating multiple accounts with the sole purpose of spreading questionable content on the platform. Automated suspicious accounts and people who seek to create havoc are barred from the platform at various stages -- at the time of registration, while messaging, and when they are reported by others, the company's executives said.

Overall, WhatsApp bans about 2 million accounts on its platform each month, a spokesperson said. To address this issue, a machine learning system uses learnings from the company's past dealings with problematic accounts and from specific scenarios engineers followed when taking down accounts, said Matt Jones, a software engineer at WhatsApp. This machine learning system has reached a level of sophistication that allows it to ban 20 percent of bad accounts at the time of registration, according to the company. Seventy-five percent of the 2 million accounts WhatsApp bans in a month are handled without human intervention or a report filed by a user, said Carl Woog, a spokesperson for WhatsApp.

11 comments

  1. I shouldn't be surprised by H3lldr0p · · Score: 1

    But I'm still surprised by the idea that someone purpose built a device that can handle multiple sim cards just so they can blast out mass messages. It's like with spam, I guess. The incredibly low rate of clicks is enough with some people.

  2. Don't ban them at registration by Anonymous Coward · · Score: 1

    Banning the account immediately only helps the abuser to switch tactics. Instead, put the account in isolation, so that the spammer wastes his time with a worthless account, and can't readily figure out why.

    1. Re:Don't ban them at registration by Gravis+Zero · · Score: 5, Interesting

      Banning the account immediately only helps the abuser to switch tactics. Instead, put the account in isolation, so that the spammer wastes his time with a worthless account, and can't readily figure out why.

      This is called a "shadow ban" and abusers become aware of them quickly because they do not get any feedback. For this reason, I propose a "Hell ban" that would effectively be a shadow ban but would employ the use of AI to simulate feedback (preferably mostly negative feedback) to keep the abuser from recognizing he's been shadow banned.

      --
      Anons need not reply. Questions end with a question mark.
    2. Re:Don't ban them at registration by ctilsie242 · · Score: 1

      I remember a bulletin board that had both shadowbanning, and an Eliza-like bot which would constantly reply to a spammer's own stuff via an account that nobody could see except the top admin and the spammers. It kept their treadmills going a while.

      What is needed are good old fashioned honeypots, and expanding teergrube functionality to suspected bots. This way, a CAPTCHA gets posted which has no right answer as one delay. Or, the account gets created and the account password doesn't work.

      I am reminded of some code I wrote ages ago, back when banning by IP or domain was minimally effective. For the MUD I had, I had code which allowed a certain IP range list to register, but every character they registered was auto-flagged where they could log on, try to spam, but the server ate the spam, then commands started getting progressively slower until the user was finally disconnected. When the user was disconnected, they got a change password dialog which would just repeat and not allow them to get past. Then, In 24 hours, the characters created from the offending IP range would be silently deleted. This kept a set of trolls spinning their wheels for weeks without affecting any of the actual players.

    3. Re:Don't ban them at registration by TuringTest · · Score: 1

      Nice idea, but what happens to the false positives?

      --
      Singularity: a belief in the "God" idea with the "demiurge" relation inverted.
    4. Re:Don't ban them at registration by Gravis+Zero · · Score: 1

      The same thing that happens to false positives currently.

      --
      Anons need not reply. Questions end with a question mark.
  3. spam bots are crazy by Anonymous Coward · · Score: 0

    As mod of a small vehicle enthusiast forum which realistically has probably ~300 active users per season ... we ban ~100 spammers a month, all year long. For a larger platform where spam bots can reach a much larger number of users, the number of spammer accounts must be INSANE.

  4. hurrah for censorship! by astrofurter · · Score: 1

    Yay censorship! Yay Faceboot for MOAR CENSORSHIP!! Yay!

  5. Cannot forword more than 5 people by gadgetsthub · · Score: 1

    In WhatsApp you can not forward the same message more than five people in our country. Sending bulk messages and creating multiple accounts with the sole purpose of spreading questionable content on the platform is restricted. If you need assistance to repair printer in windows 10 have look at https://www.canonprintersuppor...

    --
    gadgetsthub