Elon Musk Demands To Know How Many Scam Twitter Accounts There Are, But Experts Believe He’s Choosing The Wrong Route

May

15

0 comments

Tesla CEO Elon Musk sent Twitter shares surging on Friday after declaring that he would put his $44 billion takeovers of the social media company “on pause.” At the same time, he examined the number of false and scam accounts on the platform.

Musk later explained that he was loyal to the contract, but he maintained to focus on the issue of bogus accounts. He said on Twitter that his staff would perform their inquiry and challenge the data’s authenticity provided in Twitter’s most recent financial filings.

Twitter revealed in its first-quarter earnings report this year that its platform has several “fake or scam accounts,” in addition to legitimate monetizable daily active usage or users (mDAU). “We operated an internal evaluation of a sample of accounts and estimate that the average of fraudulent or spam accounts during the first quarter of 2022 constituted fewer than 5% of our mDAU,” the corporation stated.

Over the last three years, Twitter confessed to overestimating user figures by 1.4 million to 1.9 million users. “We introduced a feature in March of 2019 that lets people to connect various independent accounts together in order to simply swap between accounts,” Twitter explained. “An issue was occurred at the time, and all connected accounts were counted as mDAU as a result of operations performed via the principal account.”

While Musk’s interest is reasonable, professionals in social media, deception and statistical analysis think his suggested scheme to further research is grossly inadequate.

Here’s what Elon Musk, CEO of SpaceX and Tesla, says he’d do to figure out how many spam, fraudulent, and duplicate accounts there are on Twitter:

“To find out, my team will pick random 100 @twitter followers.” I encourage others to try the same research and see what they come up with.” In further tweets, he emphasized his strategy, saying, “Pick an account with a lot of followers,” and “Ignore the first 1000 followers, then pick every 10th.” Better ideas are welcome.”

Without proving his claim, Musk also stated that he chose 100 as the sample size for his study because that is the number Twitter uses to calculate its earnings reports.

“Any sensible random sampling process is fine. If many people independently get similar results for % of fake/spam/duplicate accounts, that will be telling. I picked 100 as the sample size number, because that is what Twitter uses to calculate <5% fake/spam/duplicate.”

When questioned if his explanation of Twitter’s technique was proper, the corporation refused to respond.

Dustin Moskovitz, a co-founder of Facebook, joined in on the debate through Twitter, saying that Musk’s method isn’t truly random, depends on small sample size, and allows space for enormous errors.

He wrote, “Also I feel like ‘doesn’t trust the Twitter team to help pull the sample’ is it’s own kind of red flag.”

BotSentinel founder and CEO Christopher Bouzy said that his company’s study indicates that 10% to 15% of Twitter accounts are likely “false,” including fakes, spammers, scammers, nefarious bots, duplicates, and “single-purpose hate accounts,” generally aim and assault people and others who mislead people on purpose.

BotSentinel utilizes a mixture of machine learning techniques and teams of human reviewers to analyze and identify inauthentic behavior on Twitter independently. Presently, the organization tracks over 2.5 million Twitter accounts, mostly from English-speaking individuals.

“I think Twitter is not realistically classifying ‘false and spam’ accounts,” Bouzy said.

He also cautions that the number of fake accounts can vary based on the issues being described on Twitter. BotSentinel observed that insincere accounts tweet more about politics, cryptocurrency, climate change, and covid than those on Twitter about non-controversial topics like cats and origami.

About the author, Awi Khan

Leave a Reply

Your email address will not be published. Required fields are marked

{"email":"Email address invalid","url":"Website address invalid","required":"Required field missing"}