Musk and Yaccarino have promoted updates to the site’s policies, such as allowing advertisers to block their posts from appearing alongside certain types of content. Advertising sales continue down by half He said this week, since Musk took control of the company in October. In part, this is because companies don’t believe much progress has been made in resolving the problem.
“Musk doesn’t live up to his promises to advertisers, and their ads appear next to truly harmful content,” said Callum Hood, director of research at the Center for Countering Digital Hate.
During Musk’s tenure, hate speech toward minorities has increased, according to the Advisory Council on Human Rights. Reports of harassment have risen and extremist content has risen, according to the Anti-Defamation League. And misinformation about the COVID-19 virus has skyrocketed, according to Media Matters.
Twitter said, after reviewing research reports, that many of the harmful content has since been evaluated and addressed, in some cases by rating, demoting or removing posts. More than 99.99% of tweet impressions, or views of a tweet, are from content that doesn’t violate the Twitter Rules, according to the company.
After publishing this story, Yacarino chirpHe described the researchers’ findings as “incorrect, misleading and outdated”.
Twitter has made a series of changes to its content security efforts under Musk, such as relaxing its rules, laying off trust and safety staff, reinstating accounts previously banned for violating platform policies and removing verification stickers on high-profile accounts that don’t want to pay for a blue tick. These changes, in addition to turning off advertisers, have alienated many users. One in four Twitter users said they are unlikely to remain on the platform next year, according to a survey by Pew Research.
Yaccarino, since starting the job in June, has spoken of a “freedom of expression, not access” strategy with brands, encouraging them to use the new controls for ads that appear next. More than 1,600 brands now use it, according to a person familiar with the matter, who asked not to be identified sharing internal data. Yaccarino has also solicited plans from third parties to improve brand controls. Meanwhile, so was Musk say Hate speech impressions have decreased.
Hood said Musk’s argument “doesn’t hold up,” citing the high level of hate speech and participation in it, according to the Advisory Council on Human Rights. During the first three months of Musk’s tenure, it contains his average daily tweets Insults against black Americans more than tripledsaid the organization, basing its research on social media analysis tool Brandwatch. From October to March, tweets referring to the LGBTQ+ community along with insults like “groomer” up 119%. Online hate often leads to real harm: reporting harassment on Twitter up 6% This year, according to the ADL.
“Musk has said over and over that hate speech has gone down on the platform, but based on our data studies, we haven’t seen that,” said Kayla Gogarty, director of research at Media Matters. “We’ve seen the opposite.”
The company told Bloomberg that Twitter’s approach to managing hate speech focuses on limiting the number of times people see it, not the amount of content itself. Twitter said impressions of hate speech content are 30% lower on average than they were before Musk’s acquisition. The company also noted that “groomer” is not an error in their policy standards, but rather a violation of the hateful behavior policy when grouped with malicious words toward a group of protected categories.
Twitter users have also reported seeing violent and sexually explicit content on the platform. Video from a mass shooting at a Texas mall earlier this year was shared publicly on Twitter hours before the company took any action; So was a video of a cat in a blender.
More than 30% of American adults who used Twitter between March and May reported viewing content they considered bad for the world, according to conducted the survey By USC Marshall Neely Social Media Index. This percentage was higher than competitors Facebook, TikTok, Instagram and Snapchat. Many users reported seeing tweets condoning or glorifying violence towards marginalized groups or explicit videos accessible to minors.
Earlier this year, researchers at Stanford Internet Observatory It found that Twitter failed to remove dozens of images of pedophilia. The team identified 128 Twitter accounts selling child sexual abuse material and 43 known cases of CSAM. “It is quite surprising that any known CSAM would appear publicly on major social media platforms,” said lead author and chief technology officer David Thiel. Twitter responded to this issue after being contacted by researchers. This year, Twitter removed 525% more accounts related to child sexual exploitation content than last year, according to the company.
Twitter has been slow to catch and remove some malicious content since Musk fired or faced resignations for nearly 75% of Twitter’s employees, including the bulk of the Trust and Safety team, which is responsible for managing responses to content reports. in the middle, only 28% of the anti-Semitic tweets reported by the Anti-Defamation League between December and January were either removed or sanctioned. The group found posts by drawing a 1% sample of all posts from the Twitter API, or API. The company said Twitter has since restricted reported tweets that were found to violate policies.
said Nadim Nashif, director of the company 7amlehThe Arab Center for the Advancement of Social Media.
During Musk’s tenure, content from extremist political groups and misinformation related to national politics has increased. QAnon related hashtags rose 91% in May compared to the previous year, with the majority of those tweets occurring in the past six months, according to research from ADL. In the first six months under Musk’s rule, Almost a quarter Covid-19-related tweets included information about unproven and untested vaccines, according to research by Media Matters, a donor-funded, left-leaning nonprofit media watchdog group. In November, Twitter removed its ban on COVID-19 misinformation.
The challenge with the Freedom of Speech rather than Access policy, said Yael Eisenstadt, chair of the ADL’s Center for Technology and Society, is “there is no way to check what isn’t already inflated.” Meanwhile, Musk himself also shared with extreme soundsand reply to Anti-Semitic conspiracy theories And Anti-trans narrativesWhich boosts those posts because it has 148 million followers.
It will become more difficult to independently verify the integrity of the Twitter platform over time. In February, Twitter Charging has begun to access its Application Programming Interface, or API. The Twitter API is used by third-party applications and researchers to analyze Tweets. While researchers previously could access millions of tweets for free to conduct research, thousands of dollars are now being paid for access to a fraction of that amount. Search Online hate speech after the 2016 presidential electionNYU’s Center on Social Media and Politics analyzed more than 750 million tweets. Today, the university would not be able to afford such research.
“Now it costs $42,000 a month to reach just 10 million tweets,” said Joshua Tucker, co-director of the NYU Center. Researchers at Stanford, Berkeley, CCDH, and ADL also said they could no longer afford access to Twitter data. CCDH is a US- and UK-based non-profit organization funded by philanthropy and donations, that aims to protect human rights online. ADL is a New York-based nonprofit, also funded through donations, that says it fights all forms of extremist hate.
To reassure audiences and advertisers, Twitter says it’s working with independent companies like Sprinklr, DoubleVerify, and Integral Ad Science to rate content on its platform. According to Twitter, the company’s brand security controls are now more than 99% effective at placing ads next to safe content.
But the damage may already have been done. The advertisers said they left Twitter over concerns about harmful content — including Musk’s own posts. Companies generally operate with smaller marketing budgets, and have more choices about where to spend their money on digital platforms. The competition for their ads will intensify very soon; For example, Meta Platforms Inc. To eventually introduce ads to its version of Twitter, Threads.
He said, “We’ve seen some advertisers withdraw their Twitter spending due to brand safety concerns.” Natasha BlumenkronVice President of Paid Social at tenueti. “For our advertisers specifically, we haven’t seen them come back to Twitter.”
The company, which is still losing money, is trying to devise business models that are alternatives to advertising. Witness the company’s premium Twitter Blue subscription, which costs $8 per month little absorption. Twitter this month began paying some Twitter Blue subscribers a share of ad revenue, based on the amount of engagement with their tweets. This rewards accounts that interact the most with Musk himself. DM