Study shows banning false info traffickers online can improve public discourse
phys.org
phys.org
As any one who’s read snopes or fact check, while the fact checks often point out actually incorrect information, some of the time, they reframe the question/narrative, assume an inaccurate scope or apply semantics in a convenient way and label something that isn’t necessarily false as not false that reflects the bias of the author.
It leads to a basic mistrust of what’s “false info” and what is not “false info” and so some people have a hard time trusting these types of studies and the basic premise that platforms should ban “false info”.
I feel Twitter’s community notes function is actually pretty good at combatting false info without creating a centralized source with the above issues.
Our entire current system of spreading information seems to allow liars who get popular on basically false outrage bait to have a huge megaphone. I'm not sure how to stop this without a centralized authority shutting it down.
Centralized authorities are immune from demagoguery? You would throw out the first amendment to protect us from liars getting megaphones?
Before social media things were generally better and we didn't have big groups like qanon or people storming the capital trying to end democracy.
Back then we did have centralized authorities doing quality control on the airwaves.
>Just look at all the false stuff elon musk has tweeted.
>allow liars who get popular on basically false outrage bait to have a huge megaphone
Do you have any kind of evidence or proof, for any of these claims? It sounds conspiratorial and emotional, especially when you say "it seems" about something that is the opposite of what I've seen in the CN program.
>I'm not sure how to stop this without a centralized authority shutting it down.
This honestly comes off as satire.
Twitter is littered with high profile false information with no community notes. So the system isn't working.
And I'm not saying both sides do it equally.
Do you have an example of this, on Snopes or another fact checking website?
[0] https://www.astralcodexten.com/p/the-media-very-rarely-lies
"We reviewed the numerous instances of Snopes' left-wing bias that we found during our June 2020 Editorial Review, such as slant. We also noted a number of times that Snopes had recently interpreted things in favor of the left, including when it "fact checked" a subjective opinion on Alexandria Ocasio-Cortez (D-NY), when it defended Gov. Andrew Cuomo by saying an accurate tweet about him was "Mostly False," and when it "fact-checked" satire from humor website The Babylon Bee (an entry Snopes then had to edit following criticism)."
-- https://www.allsides.com/news-source/snopes
They claim Snopes has a "Lean Left" bias.
Second, I find issue with the examples they did use. They say the tweet was actually true, but it doesn't seem that way to me. The tweet argued that Cuomo said that the vaccine being developed under Trump was bad, but the thing he actually said was that he thought the roll out plan would be bad. Very different.
Allsides does a systematic review. Those are a few examples from the review. You can read about their process for determining bias at https://www.allsides.com/media-bias/media-bias-rating-method.... One part of the process is an editorial review:
"The panel looks for common types of media bias such as slant, spin, sensationalism, and story choice. We review the outlet’s homepage, headlines, recent articles, photos, and other content dating as far back as six months using online archival tools like the Wayback Machine. Taking into account all perspectives, panelists individually assign a number, between -6.0 and +6.0, that they believe best represents the bias of the media outlet. These numerical ratings are then transcribed into a weighted average."
>> The tweet argued that Cuomo said that the vaccine being developed under Trump was bad, but the thing he actually said was that he thought the roll out plan would be bad.
The statement that Snopes fact checked was "Did Gov. Cuomo Say It Was 'Bad News' Pfizer Vaccine Progress Came Under Trump?".
Cuomo said "The good news is the Pfizer tests look good, and we’ll have have a vaccine shortly. The bad news is that it’s about two months before Joe Biden takes over, and that means this administration is going to be implementing a vaccine plan."
If someone is going to call the original statement "Mostly False", then yes, I would consider them to have a left bias. Cuomo is saying that it is good that we had vaccine progress, but bad that it happened before Biden was President. That sounds like an awful thing to say, and from my perspective, Snopes is covering for Cuomo who said something he would probably in retrospect like to have said in a different way.
So they would criticize someone who said it leaked out of a lab with evidence that suggested that the virus wasn't man made. However, that evidence doesn't really have much to do with:
1. Could a virus of natural origin escaped from a virus lab unmodified? 2. Could a lab that was engaged in research trying to modify a virus have accidentally let the virus escape and the research was the actual source of the virus?
To be fair, I personally don't think anyone can say for sure what really happened, but by manipulating the scope of the question to a very narrow man made virus escaped from a lab when the original speaker just said it leaked from a lab, your claim can easily be labeled as false.
We are lucky if there is hard data at all to support some decision.
But thinking about it more, we are not saying that all information has to be true or that someone can't be wrong.
I think it is also easier to identify obviously false information than it is to identify true just due to the nature of discovering things. It is easier to rule out than identify. That also helps combat the "censorship" claims.
So it seems like it would be seeing a pattern of false information. Maybe they do legitimately believe what they are saying because they fell down some rabbit hole. But if a majority of subject matter experts would say something is false and that happens over and over again for someone, than they are likely not contributing much to the platform.
I agree, it's pretty good. Except when it gets applied to one of Elon's posts and then mysteriously disappears.
If you cannot reasonably prove you acted in good faith - or cannot prove you did not knowingly make false statements it’s (and now that they are proved false you retract and remove) then you are a problem.
What that looks like in actual practice I don’t know but burying our heads in the sand and saying “it’s to difficult to stop” is not a good position to have.
I believe the term that you're looking for is "chilling" not "spillover" effect. This not a novel effect of censorship by deplatforming.
So it also doesn't take into account the potential long term toxicity of driving everyone with certain views to other platforms, where those views will be reinforced. With those platforms having the appeal of their "taboo" helping to market them.
It isn’t that it isn’t worth protecting or caring for, it’s that it disappears under the noise floor. The bad actor doesn’t have to silence the free speakers, he just needs to amplify the bullshit so nobody can hear anything that makes sense.
Are we saying that the conversation is less angry when everybody basically agrees?
For example, if there were several scientists that were debating more precise ways of calculating the circumference of Earth, the discourse would improve if flat Earth information was excluded.
diversity attractive until polarized
Every single time it's attempted, it fails.
Strong moderation in communities does make those communities better. We're in one of them. Your workplace is likely one of them. Your social sphere is one too hand curated by you. IRL we call this social norms and boundaries, but when it's online or in print people get weird about it.
Fortunately, reality allows for multiple inheritance, the Chinese state doesn't have to only be a fact-checking organization.
Easier for them, because they also have the authority to declare what is fact.
This was not just the Chinese state. Fauci said from day 1 that this was not a lab leak and anyone saying that should be discredited.
The comments I've found from Fauci about the lab leak theory are as follows:
> "If you look at the evolution of the virus in bats and what's out there now, [the scientific evidence] is very, very strongly leaning toward this could not have been artificially or deliberately manipulated … Everything about the stepwise evolution over time strongly indicates that [this virus] evolved in nature and then jumped species,"
You also have to remember what else was going on around this time. Namely, that the president was claiming to have seen evidence that it was a man-made virus from the Chinese, and suggesting it was a weapon, calling it the "Kung Flu", the "Wuhan Flu", etc. There were significant sociological impacts from this, including a dramatic increase in violence against Asian-Americans.
Context matters.
Many reports show man made in lab is more likely now.
> Fauci said from day 1 that this was not a lab leak and anyone saying that should be discredited
Let's go straight to the source of that National Review article, the House Oversight Committee's 2022 letter to Secretary Becerra [1]. If wade through the innuendo in the letter, it's pretty clear that:
Dr. Kristen Anderson sent an email to Dr. Fauci expressing her worries about whether the furin cleavage site in SARS2 was naturally evolved or a product of genetic engineering. To address Anderson's concerns, Dr. Fauci organized an online meeting via Zoom with her and a group of other specialists, which successfully dispelled Anderson's fears.
The fact is, following the Zoom conference, Anderson, along with three other participants, authored a research paper on the origins of COVID-19 that was published in the prestigious journal, Nature, in March 2020. This paper was not directed or "commissioned" by either Fauci or Collins. While Anderson did share advance copies potentially with them and others for their remarks, a common academic convention, there wasn't any attribution given for their possible input. Suggesting that Anderson wrote the paper to advance Dr. Fauci's political agenda is just (yes) conspiratorial nonsense.
Republicans in general don't like China or the concept of expertise (almost entirely because the CCP has "Communist" in the name, and expert opinions tend to refute the majority of things Republicans fight for). Dunking on China is evergreen political fodder for the right, no matter how outlandish. In 2020, (and still today, as far as I've ever read), the preponderance of scientific evidence points to the origin of covid being zoonotic.
The atmosphere of speculation by politically-motivated buffoons didn't help clarify any of this, and continues to confuse people today.
[1] https://oversight.house.gov/wp-content/uploads/2022/01/Lette...
How about the Russian disinformation campaign about Hunter's laptop and the government insistence that social media discredit and actively stop any spreading of the information. Dozens of intelligence agents said it was not credible and shouldn't be trusted. You know, the same laptop that was introduced into evidence today in Hunter's trial.
The only implementation of fact checking I’ve seen which has yet to turn corrupt is Twitter’s Birdwatch which is a much better record than these journalistic organizations we call fact checkers who feel they should control truth.
https://x.com/iAmTheWarax/status/1797732901846692117
(Yeah, @iAmTheWarax's tweet is a shitpost. It's apparently also been quoted in Newsweek as confirming the story.)
Also, there is an interesting recent study from a group at Vanderbilt University that looked at the effect that European speech regulations have in muzzling of online speech. Featured in Reason Mag.
Preventing “Torrents of Hate” or Stifling Free Expression Online? https://futurefreespeech.org/preventing-torrents-of-hate-or-...
The reality is that we're flooded with false information, which can be generated by anyone with enough resources and motivation to change how a group of people think and act. We've seen how this can corrupt democratic processes and influence the outcome of elections with the Cambridge Analytica scandal (which still continues today), and how it can be used to rally people and cause social unrest in many countries. Social media is the most effective tool for spreading political propaganda, astroturfing campaigns, and conducting any other kind of mass psychological manipulation, starting with advertising, of course.
The current state of online services is an existential threat to modern civilization, which will only grow larger with the increase of AI-generated content. Governments and companies should be doing _more_ to control this, not less. I'm not saying that I agree with mass censorship and total government control over public discourse, but surely there's some middle ground between that and the current situation.
Subreddits are the closest thing to an implementation of what I want today. Let me go to moderated subsets of the content on a platform without taking away my right to expand my view at any time I wish.
Twitter et al allow you to do some moderation with blocking and follows. But opaque recommendation engines and new accounts present challenges for both.
Users who want to curate their experience are outliers, and are probably less likely to be susceptible to misinformation to begin with.
I think the solution starts with strict moderation on behalf of companies hosting the content, accompanied by strict regulation of tech companies by governments. Along the way, passing regulation to adopt programs and initiatives that educate people about technology and critical thinking from a very early age will provide new generations with crucial skills to combat disinformation, which could hopefully eventually spread to areas of government and industry to further propel us in the right direction.
Tech approaches alone aren't the solution, and we need major social and political long-term changes as well. But strict moderation and regulation needs to happen as a start to stop the bleeding.
The problem with deplatforming and censorship is that even the savviest users are unable to discuss what they have no way to ever see.