I really like Community Notes, and hate the rest of what Twitter has become.
But... Community Notes is subject to "tampering." Elon's either removes the CNs himself from his posts, or his brigade downvote them to infinity so they don't appear on all the misinfo he posts.
Do we have any evidence that Musk has removed a CN on his own post? I've personally seen evidence to the contrary, and he makes a point of highlighting that even he gets a CN every now and then.
As the root comment noted, one of the great things about community notes on X are that the algorithm and the data it's operating on are public. If Musk were removing notes that would be trivial to prove. The fact that such claims of tampering are never accompanied by said proof should tell you all you need to know.
How would it be trivial? Can you describe in a more specific way?
The data I can find says it was last updated 9:02 PM Jan. 5, 2025 (presumably America/Chicago from my browser). That’s a >2 day window as of writing this comment.
Not throwing any accusation, just trying to understand the technicals.
If there was any manipulation of community notes in the last 2 days, how would we know?
If there’s manipulation of this data before it is published, such as ratings or notes never hitting these data files, how would we know?
Maybe, an individual could check to see their own contributions are included in updates to the published data. Is that sufficiently common such that it would get caught?
> If there was any manipulation of community notes in the last 2 days, how would we know?
You can't know until the data is published. 2 days isn't that long though. Just wait a couple more days for the next data dump, then run the algorithm and compare the results to what the X UI was showing at that time.
> If there’s manipulation of this data before it is published, such as ratings or notes never hitting these data files, how would we know?
That would be a bit more sneaky than just outright removing notes. As you noted, you'd need a user whose ratings or notes were omitted from the dump to notice and come forward. Or perhaps with careful analysis you could prove that the manipulated data could not have resulted in the allegedly removed note being shown and then later not shown, indicating something fishy happened.
Theoretically if X wanted to improve on this system, they could go even further and implement something like certificate transparency (append-only log verified by a publicly distributed merkle tree), or create an independent third party organization that users interact with to submit and rate notes, rather than that happening through X's UI. Given the threat model though, I feel like the UX and complexity trade-offs of that wouldn't be worth it. Open sourcing the data and algorithm as X has is already far more transparency than we get from any competing social media company.
If something is "removed", that logically implies that it was there at some point in the past. Technology exists which can archive the state of webpages at the current point in time. If someone is believed to do something habitually, and others wish to gather evidence of that conduct, they might prepare in advance. I assume I don't need to connect the dots here.
If something is "never added" it serves the same purpose as removal. Even more so -- there is no evidence and none was ever available. I assume I don't need to connect the dots here.
If you do not think Elon could be controlling the moderation of his own account then you are either hopelessly naive or responding in bad faith. Either way, I will not be responding again.
I don't think "CEO is able to remove community notes" is a strong mark against the community note algorithm. No system is immune to being turned off...
I am not so sure that Musk or right-wing moderators are directly to blame for the lack of published community notes. My guess: in recent months, many people (e.g., me) who are motivated to counter fake news have left Twitter for other platforms. Thus, proposed CNs are seen and upvoted by fewer people, resulting in fewer of them being shown to the public. Also, I ask myself: why should I spend time verifying or writing CNs when it does not matter - the emperor knows that he is not wearing any clothes, and he does not care.
> Elon's either removes the CNs himself from his posts, or his brigade downvote them to infinity so they don't appear on all the misinfo he posts.
I don't know if this is the case, but X is Elon's property, so he can shape it as he pleases. Assuming that X (or Facebook) is unbiased and working for your benefit is simply foolish, unless you are Musk (or Zuckerberg).
From the sources I could find quickly to refresh my memory:
> Over the weekend, Musk shared some of Roth’s past tweets and what appears to be an excerpt from his PhD thesis about Grindr, the LGBTQ social media app. Roth is quoted as saying that the app is possibly too “lewd or hook-up-oriented” for people under age 18 who are already using it, but that providers should “focus on creating safe strategies … for queer young adults” that aren’t just about hook-ups. Musk commented, “Looks like Yoel is arguing in favor of children being able to use adult services in his PhD thesis.” On Monday, the tweet had more than 60,000 likes and received 15,000 retweets.
The thesis demonstrably exists (https://uploads-ssl.webflow.com/60981d118b006454de9222b2/61d...), and it does have a roughly matching quote at the bottom of PDF page 257 (labelled page 248). The idea of businesses "crafting safe strategies" to "safely connect queer young adults" (the context is very clear that Roth refers to people under the age of 18) is very reasonably interpreted as Musk did. There are very obvious reasons why existing services advertise themselves as 18+ and attempt to enforce that, and it should be clear to everyone that any such service intended specifically for minors could not plausibly be rendered safe.
The idea that this observation constitutes an accusation of pedophilia is 100% media spin, and does not reflect Musk's words.
Ideas like Roth's are not rare on the American (or Canadian) left, especially where they intersect with LGBT etc. rights - which is how things like https://www.cbc.ca/cbcdocspov/episodes/drag-kids can come to exist and be vigorously defended. This empowers quite a bit of culture warring from the American right.
Hey, all I asked is whether Musk implied he was a pedophile. I never said the dissertation didn’t exist.
I’m sorry you took the effort to prove that point, however can you hold me responsible for that point? I was just asking a question.
Isn’t the way you interpret those sentences your issue? Not mine?
> The idea that this observation constitutes an accusation of pedophilia is 100% media spin, and does not reflect Musk's words.
It surely doesn’t reflect Musk’s exact words.
I dont know what you mean about media spin.
I do know that people took those words to imply he was a pedophile. Which is why they went out to teach him a lesson, threatening him, harassing him and hounding him. I mean, who would let a censoring pedophile be?
And it’s not Musk’s fault people did that, anyone who reads his sentence will see he clearly doesn’t mean that Roth is a pedophile.
That’s the fault of the people who can’t read correctly! Like how can I be held responsible if you didn’t understand me.
>I dont know what you mean about media spin. I do know that people took those words to imply he was a pedophile.
Media stories asserted that he made such an implication. I believe that people came to that conclusion because they saw the assertion in the media, and didn't look into it further. I don't think they could have come to that conclusion on their own from what Musk actually said - if they were thinking clearly. As you say, it was not his fault. But people very often don't think clearly when it comes to topics like this. They infer things that were not implied.
The post I replied to was accusing Musk of posting "misinfo". I responded by asking for evidence of Musk saying things that are provably untrue, because that is the standard of evidence that would be required to support such an accusation. This is not a criminal proceeding.
I don't read Musk's tweets, or really any tweets that aren't directly linked to me by friends.
That's completely irrelevant to the discussion, however. GP asserts that Musk takes actions to protect "...the misinfo he posts". The burden of proof is on GP to establish that Musk posts "misinfo[rmation]", a word which I understand in context to mean "content which is provably untrue". If you think it should mean something different, please explain why.
The obvious context is that either Meta gets out of the content moderation game voluntarily, or the incoming admin goes to war with them.
> focusing our enforcement on illegal and high-severity violations.
I imagine this will in practice determine how far they can go in the EU. Community notes, sure. No moderation? Maybe not.