More than 99% of content users and advertisers see on Twitter is healthy. And the reach of hate speech on Twitter continues to represent an extremely small fraction of the overall conversation. This past March, Twitter partnered with Sprinklr to understand, measure and reduce hate speech using its AI-based model and to further our commitment to create a brand-safe environment for our agencies and advertisers. Today we’re sharing an update that includes all of April and May 2023. Sprinklr’s independent model continues to show the reach of daily English-language hate speech impressions is even lower than Twitter’s own model estimates. Sprinklr estimates the average daily number to be 0.003% compared to Twitter’s estimate of 0.012% for the period of January 1, 2023 to May 31, 2023. Additionally, we estimate hate speech impressions are 30 percent lower on average vs. pre-acquisition. In June we significantly expanded our Freedom of Speech, Not Reach policy enforcement, which dramatically reduces impressions on harmful content. And all indicators for June and July show sustained progress. Our work is ongoing and we’re proud of the progress we’re making in maintaining a healthy global town square that is open for everyone to be themselves.
More than 99% of content users and advertisers see on Twitter is healthy. And the reach of hate speech on Twitter continues to represent an extremely small fraction of the overall conversation. This past March, Twitter partnered with Sprinklr to understand, measure and reduce hate speech using its AI-based model and to further our commitment to create a brand-safe environment for our agencies and advertisers. Today we’re sharing an update that includes all of April and May 2023. Sprinklr’s independent model continues to show the reach of daily English-language hate speech impressions is even lower than Twitter’s own model estimates. Sprinklr estimates the average daily number to be 0.003% compared to Twitter’s estimate of 0.012% for the period of January 1, 2023 to May 31, 2023. Additionally, we estimate hate speech impressions are 30 percent lower on average vs. pre-acquisition. In June we significantly expanded our Freedom of Speech, Not Reach policy enforcement, which dramatically reduces impressions on harmful content. And all indicators for June and July show sustained progress. Our work is ongoing and we’re proud of the progress we’re making in maintaining a healthy global town square that is open for everyone to be themselves.
Since the algorithm went public with tweaks….I keep getting recommended misogynistic, sexist, racist, homophobic and inflammatory tweets from Republicans. My timeline used to be movies, tv shows, news, people I’m following & lower tier users. When I quote something, I’d start seeing more tweets about that topic — hence topics were loved by many. But that is not the case. I’m tired of blocking people. Surely, that should be looked at? seems like Daddy Elon is pushing only conservative views?? I don’t wanna be hate crime’d daily.
@TwitterSafety Twitter does a good job of hiding hate speech, though they give you the option of viewing it if you want
@TwitterSafety Today they tried to steal 5 scams from me, 4 had a blue check and 1 gold check, this week there were 18 blue check scams and 4 gold checks. and you want us to believe what you say. When twitter finds a blue or gold check to be a scam, what does it do?
@TwitterSafety I had to block the same dude on five alt accounts the other day for being a racist and threatening to kill other people. Every time he came back with an alt. Stop playing
@TwitterSafety The 1,000,000 thirst bot accounts that like people’s tweets disagree. This is annoying as hell. Why is Twitter so damn horny all of the time?
@TwitterSafety ANY control over any speech is censorship and not “Free Speech”.
@TwitterSafety Oh so the new word is freedom of speech not reach, and there’s still no solid definition of what is hate speech. @elonmusk are you sure you’re not being woke here?
@TwitterSafety Why do we need AI to police the platform? Isn't this Sprinklr thing a holdover from Twitter 1.0? Does the feedback from this thing result in people getting shadow banned?
@TwitterSafety There is no such thing as hate speech. There is only strings of words that you don't personally like.