Comment on Biden signs TikTok “ban” bill into law, starting the clock for ByteDance to divest it

• The_Sasswagon@beehaw.org ⁨2⁩ ⁨months⁩ ago

  If I’m reading that right, that could also say that Instagram is suppressing anti-israel content? It’s just saying that in comparison to Instagram tiktok is showing more x, y, z. But Instagram is absolutely not a neutral point to measure from.

  For starters there’s different demographics on each one, but I’m sure you could adjust for that, maybe the study did. But I don’t think you can adjust for the impact the US government has on Meta. I don’t believe for an instant that some US agency isn’t manipulating algorithms or requiring certain tweaks to steer discourse just like they did with US news outlets.

  source

  • Dark_Arc@social.packetloss.gg ⁨2⁩ ⁨months⁩ ago

    So, I took another look at the report, they did do this sort of statistical bias correction. See “U.S. Politics” page 8 networkcontagion.us/…/A-Tik-Tok-ing-Timebomb_12.2…

    source
  • Dark_Arc@social.packetloss.gg ⁨2⁩ ⁨months⁩ ago

    Hm… I agree that Instagram is not a neutral source. I also agree that there are going to be some biases imposed by the user base.

    I don’t believe the US government plays a major role in Meta’s content moderation behavior. Meta if anything has shown a reluctance towards any political or news content in recent years. That’s not to say the US government doesn’t have influence but their influence is (from what I’ve seen) oriented around fighting disinformation and threats of violence … not cherry-picking the discussion of subject matter. I think there would’ve been a pretty significant leak out of Meta by now if there really was a strong political bias or government influence in content moderation.

    I don’t think any of these lines particularly fall along political lines within the US either. There are people on the left and right taking different sides on virtually all of the topics with statistical divergence.

    source