Why can’t social networks stop fake accounts?

Social media companies say they deleted an additional billion fake accounts every few months. So, a 21-year-old delivery driver in Pennsylvania impersonated a member of the Trump family on Twitter for nearly a year, eventually tricking the president.

The answer has to do with the size of social networks, the complexity of catching fake, and the business incentives of the companies that run the site.

Facebook said it blocked 4.5 billion accounts in the first nine months of this year and captured more than 99% of those accounts before users flagged them. That number of accounts (equivalent to almost 60% of the world’s population) can be daunting. It is also inflated.

Most of these accounts were so-called bots, or automated accounts that were often created together by software programs. Bots have been used for years to artificially amplify certain posts and topics so that they can be seen by more people.

In recent years, Facebook, Twitter and other tech companies have been very good at capturing bots. They often use software to find and block them during the registration process by looking for digital evidence that suggests that their accounts are automated.

As Facebook catches more bots, Facebook also reports a huge statistic on the number of fake accounts to delete. These numbers have given the company a lot of positive headlines, but “it’s not really used that much internally,” said Alex Stamos, Facebook’s former chief information security officer who left the company in 2018. Says. , Because it doesn’t cost anything to try. “

In other words, one person can create a software program that tries to create millions of Facebook accounts, and when Facebook software blocks those bots, the aggregate of deleted fake will spike.

Facebook admits that these statistics aren’t very useful. “The number of fake accounts has been significantly distorted by a simple attack,” said Alex Schultz, Facebook’s vice president of analytics, last year. The prevalence of fake accounts is a clearer indicator, he said. And that shows that the company still has big problems. Facebook estimates that 5% of profiles are fake, or more than 90 million, despite deleting billions of accounts, which hasn’t been out for over a year.

Social media companies are much more struggling to create fake manually created accounts, that is, people who sit in front of a computer or tap a phone.

Such fakes do not have the same clear digital signage as bots. Instead, corporate software needs to look for other clues, such as the account sending the same message to multiple strangers. However, that approach is incomplete and works better for certain types of fakes.

This explains in part why Pennsylvania delivery driver Josh Hall was able to repeatedly impersonate President Trump’s relatives on Twitter and attract tens of thousands of followers before the company noticed.

Manual fakes can be more harmful than bots because they look more reliable. Political agents use such disinformation to spread disinformation and conspiracy theories, and fraudsters use them to deceive people. Criminals are tricking people into giving money on social media under the guise of celebrities, soldiers, and even Mark Zuckerberg.

Twitter’s efforts to catch fraudster accounts are complicated by policies that allow parody accounts. The company needs to clearly label their parody accounts.

Facebook is also struggling with accounts disguised as public figures, but regular reviews by the New York Times suggest that the company is good at removing them. Facebook-owned Instagram hasn’t made much progress.

One way to combat counterfeiting is to request more documentation to create an account. Companies often started asking for phone numbers, but they hate making it harder for people to join their site. Their business is premised on adding more users and allowing them to sell and display more ads. In addition, Twitter especially appreciates user anonymity. The company said it would allow dissidents to speak to authoritarian governments.

Therefore, companies rely on flagging users to reduce the number of suspicious accounts that need to be reviewed. This strategy is much more efficient and cost-effective for businesses. It also means that as fake accounts get more attention, they are more likely to be flagged for more details.

Still, companies can still take some time to act. Hall gained 77,000 followers disguised as President Trump’s brother and 34,000 followers disguised as President’s 14-year-old son before deleting the account Twitter was using to spread the conspiracy theory. Did. From 2015 to 2017, Russian government workers disguised themselves as Tennessee Republicans on Twitter, attracting 150,000 followers, including senior members of the Trump administration, and racist and xenophobic, according to a federal survey. I posted the message.

“We strive to quickly and consistently deal with spoofing rule violations, especially when people are trying to disseminate false information,” a Twitter spokesman said in a statement.

Still, most fakes can’t attract many followers. Stamos argued that fraudster accounts, which most people are unaware of, have little impact. “It’s pretty Zen, but if no one is following a fake account, does it exist?” He said.

As tech companies face so many threats, Stamos has to make difficult decisions about which issues to tackle, and sometimes the tricky task of eradicating each fake account is worth it. Said not.

“Companies usually put their efforts behind what they can show not only to look bad, but also to be the worst,” he said. “How do you always apply finite resources to the problem that is actually harming you?”

Why can’t social networks stop fake accounts?

Source link Why can’t social networks stop fake accounts?

Back to top button