U.S. Markets closed
  • S&P Futures

    -17.00 (-0.49%)
  • Dow Futures

    -135.00 (-0.48%)
  • Nasdaq Futures

    -42.25 (-0.36%)
  • Russell 2000 Futures

    -12.70 (-0.78%)
  • Crude Oil

    -0.59 (-1.48%)
  • Gold

    -4.10 (-0.22%)
  • Silver

    -0.23 (-0.95%)

    -0.0020 (-0.1659%)
  • 10-Yr Bond

    -0.0070 (-0.83%)
  • Vix

    -0.56 (-1.99%)

    -0.0005 (-0.0404%)

    +0.1400 (+0.1337%)

    +93.29 (+0.72%)
  • CMC Crypto 200

    +2.35 (+0.90%)
  • FTSE 100

    +74.63 (+1.29%)
  • Nikkei 225

    +3.50 (+0.01%)

Twitter suspends accounts for posing as Black Trump supporters

Yahoo Finance's On The Move discuss Twitter's move to suspend accounts that posed as Black Trump supporters.

Video Transcript

ADAM SHAPIRO: These accounts have had people say, quote, "yes, I'm Black, and I'm voting for Trump." They got more than 250,000 retweets, but they appear to be actually fraudulent accounts. Rick Newman, you first, and then let's bring Melody Hahm in on this.

RICK NEWMAN: This is an old trick at this point. I mean, we saw this in 2016. But a lot of the social media companies were not taking these accounts down because no one had ever actually put together what the overall strategy is here. Twitter did not say who's operating these accounts. So in 2016, the Russians did operate some accounts like this.

It's also possible, since these were accounts in favor of Trump, that what they're trying to do is they're making-- trying to make it seem that his support among Black voters is higher than it actually is and perhaps persuade some Black voters not to vote for Joe Biden. Back in 2016, these kind of went rampant, and nobody really policed them. And they're just doing a better job of figuring out what's going on now.

You can tell, I'm sure Twitter has all kinds of tools for detecting fake accounts that we don't have. But even out here, just among us, you can tell a fake Twitter account sometimes just by downloading the picture that's in the profile and then doing a reverse image search to figure out, is that a picture that is kind of all over the internet on social media profiles under a bunch of different names? That's one tell.

So Twitter's on it, and we're probably going to see a lot more of this in the last couple weeks of the election.

MELODY HAHM: I mean, a quick edit feature, right, to what you just said, Rick. Twitter actually was not the first to proactively probe this, right? It unfortunately comes from a lot of academic researchers. In this case, it was a group at Clemson University that pointed this out and basically did a longitudinal study trying to track how long this has been going on, especially because some of the accounts that we're talking about that have only recently been suspended, they've been in existence since 2017.

So after Trump got elected, they sort of took a bunch of these handles and have been sitting on them and now tweeting for the first time, perhaps, in early October. So it's pretty fascinating to see that social media companies, their stance still very much, you know, I don't want to let them off the hook, right, because their stance very much, even if they are able to identify some of these things early on, they would rather not speak to it until it's called out.

So the fact that that is still the paradigm of the way these social media companies that leverage so much power and have these really pervasive and ubiquitous platforms are still waiting for someone to point it out, I think that's an inherent flaw in the system. I do also want to say this idea of a Blexit, which is what Black voters leaving the Democratic Party, Candace Owens perhaps the most prominent figure of that movement, has been a real phenomenon, right?

There is a certain very niche cohort that is part of that group. But to Rick's earlier point, this ability for bots, for random people, whether they're Russian trolls, whether they're White conservatives in their homes, to adjust easily quickly whip up a bunch of accounts because there's no identity verification on the platform, and that's the same for most social media platforms, is turning out to be quite a troublesome predicament, right? The AI system and algorithms and the tools that we're alluding to cannot catch all of these because there are some sort of crazy people out there who say whatever they mean.

And it may read as though it's a bot, I think the best way to trace a lot of these things, to Rick's point, is to kind of reverse image search. Because we did see some of the shoddily done jobs, right, where the photo was actually just an image that says Black man walking, right? So unfortunately, a lot of ways to detect this.

But it does take a lot of work. This is not an easy sort of fix that, you know, one small department within a social media platform has been able to do.

RICK NEWMAN: Hey, Mel, I think you're right. It does stand to reason that Twitter and Facebook both, and YouTube, we should add, could be more aggressive about finding these. I mean, come on, they do have better tools. But it's also true that everybody is more aware of it now. And we're sort of crowdsourcing the policing of these sites.

So, I mean, yes, you'd prefer that Twitter were finding this before it became kind of rampant. But it's also a good thing that academics and journalists and others are doing it. And, you know, the most important thing for anybody just thinking how to tell is just be skeptical. I mean, if you see these-- if you see these feeds that just seem like, really, does that make sense?

You know, look at the bio and see where you can figure. And if it looks fishy, it's fishy. You know, one thing, I've reported on this back in 2016 and '17, and one way you detect-- or you defeat the AI is you have older accounts that have been in existence for some time. And you can actually buy these accounts on the internet for, like, you know, a fraction of a-- a tiny fraction of a Bitcoin, and it's usually Bitcoin, so it's untraceable.

And there are accounts that have been around for 10 years on Facebook and Twitter that people buy because they seem more real than ones that were just started, you know, a month before the election. So everybody should be super skeptical.

MELODY HAHM: Yeah, quickly, this idea of digital blackface, that's what researchers are calling it, you know, we talk so much about cultural appropriation. In this case, it is very much a digital blackface, right? There are a lot of people who are not Black who are pretending to be Black to try to sway the vote.

But I do want to point out, in addition to those accounts, there were a couple other. There was one pretending to be Erica Kious, who's Nancy Pelosi's hairstylist. We saw all the controversy there. A few pretending to be White police officers in Philadelphia. So even just zooming in and focusing on one, sort of, demographic clearly will not root out the problem, right? This is a much long-standing pervasive issue that will be a public private partnership, sort of, endeavor to really get this weeded out.