It’s been nearly 6 months since Elon Musk took control of Twitter and began making changes.
The changes were met with headline after headline warning that Musk was fostering hate online.
Now Elon Musk is celebrating after this independent assessment put a positive light on his fight for free speech.
Critics said Musk was putting up a “Bat Signal to every kind of racist…”
After Elon Musk purchased Twitter in October of last year, the corporate-controlled press went into a full meltdown. They portrayed his goal of making speech free again as his way of bringing hate speech back to the center of the conversation.
Despite their attacks, Musk forged ahead to loosen guidelines on the platform and welcome back thousands of accounts that had been suspended under the veneer of safety. The new owner and CEO of Twitter promised that none of these actions would result in hate speech.
Leftist activists pushed back at Musk. The New York Times published an article in December titled Hate Speech’s Rise on Twitter is Unprecedented. The article quoted the CEO of the Center for Countering Digital Hate, Imran Ahmed, who called Musk’s purchase a “Bat Signal to every kind of racist…”
Now Musk’s “absolutist” view on free speech is being recognized by industry leaders as a net positive.
Tech company evaluates Musk’s Twitter with “AI-based Toxicity Model”
Sprinklr is a company that was built around interactions on social media and compiling user behavior for their marketing partners. They’ve been around since 2019 and have worked with several major companies to scale their social media efforts worldwide.
They’ve now used what they call their “AI-based Toxicity Model” to determine the level of hate and “toxicity” on Twitter since Musk’s takeover. They analyze available data and use AI software to measure angry and divisive language online.
Sprinklr says that their “AI-powered model found that the reach of hate speech on Twitter is even lower than [their] own model quantified.” They measured “hate speech impressions” rather than the number of “Tweets containing slurs.”
They found that “hate speech receives 67 percent fewer impressions per Tweet than non-toxic slur Tweets.”
Final report says hate speech is getting less reach
Musk said that their evaluation of the overall reach of toxic language was “a critical distinction” and he applauded their efforts see through the minutia. He raised the concern that “a single person” could create thousands of bot accounts that are “tweeting slurs.”
Sprinklr’s approach avoids giving these people credit for spreading hateful slurs. As Musk pointed out, “what matters is whether those Tweets are actually shown to real users.”
The company also evaluated whether all Tweets containing slurs were actually hate speech. They broke out the full context of posts and determined whether the slur words were accompanied by toxic language. They said that toxic Tweets were much less likely to gain meaningful reach.
What’s more, only about 15 percent of Tweets containing a slur also contained toxic language. The other 85 percent of posts containing slurs were written in “non-toxic contexts.” They suggested that many of these words had been reclaimed by the group the word was used to describe.
Musk’s efforts to secure free speech online seems to be going just as he planned.
Patriot Political will keep you up-to-date on any developments to this ongoing story.