- This Twitter account is parodying Amazon customer service Thursday 8:52 PM
- How to live stream Oleksandr Gvozdyk vs. Artur Beterbiev Thursday 7:00 PM
- Shaggy says an online scammer is impersonating him Thursday 6:53 PM
- IRL Barbie Malibu Dreamhouse available to rent on Airbnb Thursday 6:16 PM
- Men’s Humor trolled for unknowingly tweeting Grindr conversation Thursday 5:29 PM
- How to stream Dominick Reyes vs. Chris Weidman Thursday 5:00 PM
- Jennifer Aniston had a finsta before officially joining Instagram Thursday 4:35 PM
- Facebook denies moderating comments under Zuckerberg’s big free speech live stream Thursday 2:38 PM
- ‘My headphones’ meme proves our music is sadder than we look Thursday 1:53 PM
- ‘Time for an upgrade’ meme shows Kamala Harris’ team is too online Thursday 1:35 PM
- Prison guards reportedly mocked trans inmates in private Facebook groups Thursday 1:33 PM
- Gradient is the new celebrity look-alike app winning over influencers Thursday 12:46 PM
- Trolls accuse cosplayer of ‘appropriating’ Joker culture (updated) Thursday 12:28 PM
- Every Studio Ghibli movie will stream exclusively on HBO Max Thursday 12:24 PM
- ‘Stranger Things’ season 3 saw its highest viewer numbers yet Thursday 12:01 PM
To better fight spam and spambots on its platform, Twitter is making changes to its account creation process. Now, those trying to make a new account will need to verify it with either a phone number or email address. It’s another much-needed step in the app’s fight to make its platform friendlier, safer, and more healthy.
Twitter detailed this latest round of updates in a blog post this week. It is implementing four new ways to fight spam on its platform: Reducing the visibility of suspicious accounts in tweet and account metrics (that is, your follower number or the number of retweets a post gets); updating the signup process to require a phone number or email address for identity confirmation; auditing existing accounts for signs that they were created through automation; and a broadening of Twitter’s existing malicious behavior detection systems. High-volume tweeting with the same hashtag, repeatedly using the same @ handle without replies, and other activities will trigger the latter.
Twitter has already increased its use of machine learning to proactively detect and prevent spam and malicious automation from operating on its site.
Thanks to this system, as of May, Twitter now identifies more than 9.9 million potential spam accounts each week, up from 3.2 million in September. By spotting more potential problems it’s also removing more accounts for spam-related violations. This has had a positive effect on the number of spam reports filed to Twitter: This number is down from 25,000 reports per day in March to 17,000 per day in May. Twitter is also suspending more spammy apps for violating its guidelines—an average of 49,000 per month in April and May.
A look at those numbers provides insight into the tremendous scope of Twitter’s spam problem and some explanation as to why it’s taken the company so long to figure out effective ways to combat the issue.
Twitter’s latest techniques should help combat some of the spam-related issues that cropped up in the 2016 elections, and just in time—the 2018 midterms are around the corner.
Christina Bonnington is a tech reporter who specializes in consumer gadgets, apps, and the trends shaping the technology industry. Her work has also appeared in Gizmodo, Wired, Refinery29, Slate, Bicycling, and Outside Magazine. She is based in the San Francisco Bay Area and has a background in electrical engineering.