A recent epidemic of spam bot accounts scamming crypto followers on Twitter is having unintended consequences — the real people they’re impersonating are getting banned instead. And implementing a solution is going to be a lot harder than you think.
Subscribe to the Bitsonline YouTube channel for more great interviews featuring industry insiders & experts
Background to Crypto Twitter Scam Bans
By impersonating and then subtweeting various personalities in the community, these bots have been able to steal thousands of dollars in cryptocurrency, and are rightly reviled.
The problem is that a certain percentage of the same people who misunderstand the bots’ provenance and send them money, are reporting the genuine accounts for scamming them. This understandably causes some issues:
Sorry, guys. @TwitterSupport have permanently banned our @krakensupport account for "rules" against repeatedly warning you about the unmitigated scams in the replies. Looks like you're on your own now. pic.twitter.com/KnR34yzds9
— Kraken Exchange (@krakenfx) March 6, 2018
Twitter bots taking legitimate accounts down with them pic.twitter.com/i6LqpVeO2N
— Neeraj K. Agrawal (@NeerajKA) March 6, 2018
Why Isn’t Twitter Banning the Spam Bot Accounts Instead?
For all we know, they are — and at a faster rate than they’re banning the victims of impersonation. The problem is that there’s a lot more bots than real people, and the spammers behind them can always just create more.
To get a more technical perspective, we got in touch with the architect of the “that’s a scam” bot, which identifies these spammers and flags them accordingly. It uses a combination of follower ratios, homoglyph (similar looking character) analysis, and age/number of tweets to flag would-be impersonators.
When asked why he could so easily identify and tag false accounts when Twitter was unable to, he said it was a matter of logistics:
“I imagine it’s actually quite difficult from an engineering perspective … Twitter [has] a massive platform and would need to monitor 6,000 tweets per second to combat spam. I can only guess they have a dedicated team just for that.”
As easy as it is to blame the social media network for the recent string of bans, the fact is, there’s no good way to automate spam control on platforms this large without false positives, and pressure from media and elsewhere for Twitter to tighten up controls has only increased the proportion of misfires in their system.
While at first glance this string of bannings may look like discrimination or censorship on Twitter’s part, the sadder truth is that Twitter probably hasn’t even noticed, much less moved on, fixing this problem.
How can Twitter deal with the spam/scam epidemic? Please share your thoughts in the comments.
Images via Twitter, Bitsonline