One thing I considered with bots like these is that it doesn't matter which side of an issue they are pushing, the end result is just to further divide the 2 sides. So I think that's the goal.
The issue they push with the bots may not be, and likely isn't, their actual position on the issue.
It's part of the Russian/Putin playbook. Their mission isn't to promote any particular interests within America or other countries, as much as to maximize divisiveness and weaken democracy as much as possible so that Russia is stronger and less threatened in comparison.
You are being divided internally by a domestic force. BlackLivesMatter vs everyone, putting male against female, isolating individual by destroying nuclear family. Putin should be last of your concerns.
yes, it does not, but over last 4 years a lot of things happened that aligned with the content. Look at even the latest riots. Either the author is some kind of fortuneteller or there is an active effort toward is and unfortunately our leadership is helping it happen. I don't believe in coincidences.
Twitter is so easy to game. I once wrote a script that auto-likes everything on the page and ran into no throttling by Twitter - It simply let me rack up an enormous like count nearing 1 million random likes. I got lots of followers as a result because they thought my like was a genuine human interaction and I 'cared' for them in some way. Far from it.
I think Twitter needs to add an option which marks an account as a bot or non-human account so that people can gauge instantly what the account's real motivations are. Most actual bots will ignore that setting and pretend to be human however, so the responsibility is on Twitter to weed out automated accounts that are obvious attempts to game the platform.
Starting a sentence with "so?" implies that you don't care that X position has a lot of bots behind it. But then your followup question implies that you would care if position Y has a lot of bots behind it. Shouldn't this be a "that's bad, I wonder how bad it is on the other side" kind of situation? Why are you defending this?
Probably some, and I’d also welcome the research. But when 80-90% of the population support lockdown there’s less need for bots. Apple pie is less popular than ‘continue lockdown for as long as necessary’.
I'm cynical enough to think you can almost certainly say the same of the reverse, too. :(
That said, with reading this, if you are thinking of skipping it. Does somewhat confirm that most of the not posts are on conspiracy theories. Which, I guess is not that surprising. Even if the conspiracies are.
I often wonder when we are going to see more left wing reactions to this sort of behavior. Seems the right, based on my limited information, seems to be more aggressive about these disinformation campaigns.
Also it stands to reason that this particular misinformation campaign is being sponsored by a foreign state. I think that this would be an effective weapon to use against the United States. Sort of like biological warfare by proxy of disinformation.
Why would the use of bots to push a narrative be restricted to “the right”? Isn’t it more likely that 50% of all posts on divisive topics are made by bots? The talking points for both sides seem imminently scriptable.
Perhaps we should all step back and evaluate whether our internal thought processes are being shaped by bad actors via an appeal to our innate tribalism? Notice how easily you fall into the script: 1) it’s the other team only 2) my team should do something to fight back 3) maybe it’s sponsored by a foreign state.
I have quite a “purple feed” on FB. I have to say that I see a lot of utterly ridiculous content created ostensibly supporting both blue and red causes/topics.
If you primarily think one side is crafting ridiculously bad content and it’s therefore a misinformation campaign from a foreign state, I think it’s wise to consider how much of that is personal bias and if you’re perhaps “giving a pass” to stupid content that at least aligns with your views.
“Well, they meant well, so it’s probably just a right-minded American who made a simple off-by-factor-of-a-million arithmetic error and no one spent 50ms doing a basic sanity check. No one’s perfect.”
Or “Look, maybe I think Obama did an Ok job, but I saw all that content about he wasn’t born here, and there has to be something to it if so many people are talking about it. The government hides things; the truth is out there, ya know? Oh, and Epstein didn’t kill himself.”
It's telling that I bring this up and this is the response I get. Maybe there is a possibility that it is a state sponsored disinformation campaign? There is overwhelming evidence that disinformation is being spread by several state sponsored groups.
Again, I said from my "limited view", as it's impossible to have all the information. It seems TO ME that there is quite a bit more targeted misinformation coming from the right.
I would love to be corrected. However, I don't think saying "I got a Facebook feed with lots of stuff stuff coming out of it", and "you're stupid because of xyz I assume about you for no reason whatsoever" qualifies in this regard.
If “I think it’s wise to consider how much of that is personal bias” translates directly to “you’re stupid because of xyz I assume about you”, then it seems like we’re not effectively communicating.
I’m sorry for the portion of that where the fault is mine.
Maybe you're right. Your tone, and granted this it the Internet where tone is implied by the reader, seemed very presumptuous. I think ultimately the first part of my post was an, admittedly, biased opinion. So yeah, I am probably wrong about that. Though, I would really like to see some data on this.
However, I am concerned about a disinformation campaign which could, theoretically, be sponsored by a foreign state to get people to ignore the very real threat of coronations.
Sorry for the tone. I tend towards a little formal perhaps, but it’s intended to be optimized for precision not presumption.
I am also concerned about foreign influence, but I also observe that we meddle in other countries’ affairs as well, so it’s just a fact of life and social media changes it but didn’t start it. I’m more concerned with a population who isn’t thinking critically and with the right balance of short and long term (and is therefore especially susceptible to influence), but whether the influence is foreign or domestic doesn’t matter as much to me as the effect of the influence.
"Please don't post insinuations about astroturfing, shilling, brigading, foreign agents and the like. It degrades discussion and is usually mistaken. If you're worried about abuse, email us and we'll look at the data."
I think this is a toxic rule to have. Baseless claims don't help anything, but telling people not to talk about something because it might make the site look bad is gross.
It would be much better to show useful statistics to convince people that bots are kept to a minimum.
It more than "looks bad", it actively stifles conversation and keeps it at probably around the "ad hominem" tier of disagreement (http://www.paulgraham.com/disagree.html).
I think even if you were to present statistics of some kind, some people might not consider it enough and ask for some different contrived set, on and on, forever.
If you think something's really wrong, you email the mods about it. Otherwise you get in this exact meta-discussion land about what is astroturfing, what are accusations, and so on. And then you get threads that distract from meaningful conversations, on and on, forever.
Maybe we could all just try to discuss the article and related comments by refuting/agreeing with their central points. It would be less boring than seeing the same metadiscussion and flailing forever.
(In that vein, I won't be replying to this comment further)
> It more than "looks bad", it actively stifles conversation and keeps it at probably around the "ad hominem" tier of disagreement
If it happens a lot and without a given reason it certainly does, but it isn't such a black and white issue that it has to be stopped at all costs. When someone points out very suspicious patterns over multiple names, comments, similar threads that are more likely to have astroturfing or bots etc. and mods pop in to say not to talk about astroturfing and bots because it doesn't happen, that's not helping keep the site to a higher standard.
I don't doubt it is a priority and people are working on it, and I don't think the reply here that has no examples is worthwhile, but an all or nothing attitude doesn't help, because it is the reality of the internet.
Do you have any examples of this? Have there been any significant instances of this? I’m not saying that the moderation team is not good, but we should be allowed to question and discuss the possibility that they are not.
It's also not so much about trying not to make "the site look bad", it's more that it generally leads to baseless witchhunts and flame wars.
HN has an advantage over Twitter and other social networks: posts are supposed to be insightful instead of merely pictures of cats or people doing funny tricks. As a result we can criticize posts based on their intrinsic values instead of having to go meta and wonder about the intentions of the person posting them.
If a bot manages to posts dozens of interesting stories that make it up to the top of the front page every day I say we keep it...
Not necessarily promoting things, but I'm 100% sure there are bots crawling the comments for keywords (product/project mentions) so that the creators/employees of said product/project can show up in the comments to answer questions if need be.
Also, from time to time I see accusations that anti-CCP comments/articles are flagged to death by bots, but I've never witnessed this first-hand.
The voting and posting activity on threads heavily critical of the chinese government have seemed very suspect to me as well. Many of the names defending the government having post histories of only government defense and "what about other country" while votes end up going through cycles of rising gradually and being downvoted all at once.
If you spend enough time in those threads to notice patterns, and even look through people's posting history, you're essentially acting like a bot crawling the site.
If you then downvote those posters you disagree with, it creates a very suspicious voting pattern on their end.
Could you distinguish your own behavior from that of a bot (or a bot-assisted human) if you didn't know it was just you in front of your screen?
That makes zero sense. I'm not dropping comments that have already been upvoted into negatives in seconds and I don't have a user name of random letters a few months old with only 4 or 5 comments, all defending the chinese government in heavily commented threads about a story critical of the chinese government.
Of course you're not dropping comments from +3 to -4 (or whatever) on your own, but it only takes six other people who share your opinion to do that. As for people with only 4 or 5 comments, they're not going to have enough karma to downvote anyone.
Your first paragraph application of bots is 95+% positive I’d think and I’m 100% supportive of the use of crawlers for that and then hope the humans who pick it up act positively.
(I’m not even sure I’d consider something that was itself read-only in nature to be a “bot” but I suppose it technically is.)
If you look at most CCP-related stories posted here they almost immediately devolve into intense flame-wars where any nuance is discarded. The official reason for these posts to be killed is because they trip all sorts of anti-flame killswitches and frankly that doesn't seem far fetched at all to me. These threads are usually an embarrassment for this community honestly.
Came here to say the same thing. When I see these submissions – pro or anti-CCP – and I see the discussion that results, I'm inclined to flag them. Such stories are not good material for HN, and the submission guidelines make that clear. I thought this was precisely what flagging of submissions was for.
It's also worth noting that some of the worst stories that have resulted in the most division have been ideologically driven. I recall a libertarian think tank that had a number of accounts that apparently existed purely to make submissions for those sites. I'm sure there are other groups that are at it too, I just found that one to be particularly blatant.
Regardless of your political views, the submission of such stories seems to inevitably result in a flame-war that seems only to serve to degrade this community's technologically-orientated mutual respect.
It's pretty crazy that a few researchers can identify this in a short month or two of work.
It's even crazier that Twitter can't do that research themselves and remove those bots.
Obviously the big debate is censorship. It feels like there could be a tech solution to this, where conversations aren't shaped by volume, but I guess that is what Twitter is anyway.
How many real people is it OK to squash when squashing bots? How confident can you be when banning accounts?
It’s a lot harder problem then it might sound on the surface. And realize if you find bots 100%, they will change for round 2 and the game starts over.
> Just put in a super basic arbitration process that is costly to game for "non-real" people
I think your "just" is hand-waving away a lot of complexity. A person deployed these bots, and if a bot account gets put into an arbitration process, that person can go through the process themselves to pass the arbitration, then set the bot running again.
Sacrificing a few genuine accounts to save the sanity of millions and protect democracy is not much of an issue. The question is how does one revive such accounts without too much hassle while not reviving bots.
Easy: solve 5 captchas (all known captchas of course, not 2 knowns and 3 unknowns) a day for 2 days and you get authorized again. Next violation, solve 5 captchas per day for a week. Next one, 5 per day for a month.
That could work indeed. Annoying but necessary. Tho what does one do about real bots - people in spam farms that would manually
login, post and logout.
Maybe they don't need to ban accounts, but instead put up a "bot check", sort of like their new editorial process? Something along the lines of "This account has been identified as likely being a bot. Read more." To remedy a false positive, the human could provide Twitter with a an official ID to remove the warning.
Worse, Twitter can't be trusted to keep the IDs safe. They've suffered many data breaches in the past.
And it wouldn't even stop the bots, because Twitter isn't limited to Americans, so bot makers could easily send fake official IDs from someplace Twitter can't check. Or send stolen offical IDs from other data breaches.
I bet there are some ThisIDDoesNotExist websites that bots use to autogenerate vast majority of fake ids, probably by using ThisPersonDoesNotExist website.
You can generate realistc photos of IDs from countries that Twitter cannot positively verify those bc in USA there are ways to check authentity of your document. But IDs from Russia or Ukraine - no way.
> To remedy a false positive, the human could provide Twitter with a an official ID to remove the warning.
That would work only against casual bots. The big nation-sponsored botnets won't have issues with providing genuine government issued identities for bots. Which kind of defeats the whole purpose and wastes a lot of resources twitter would like to use elsewhere.
They already do that. Fun fact, I was never able to make and use a twitter account even after verifying it with my phone and email. Apparently I was a bot from the start, before even posting anything.
All this bot stuff/millions of fake accounts amazes me.
Whenever I sign up at Twitter (entirely because some companies do support there) the account is flagged within a few minutes for "suspicious activity", such as making a coffee and coming back to the PC or searching for the company I want to contact.
There are a lot of statistics that can be analysed to indicate a bot. If you start with extreme cases and catch 'real' people it might be for the best.
Think about an account that is active 24/7 every day all year, or one that fires off multiple actions per second for long periods of time.
Beyond that, having a limited vocabulary, repetitive phrases, consistently accessing the same thing every few minutes, etc. are all low hanging fruit.
You’d ban half of Twitter. It’s pretty much impossible to differentiate an actual “person typing like a bot” from a legitimate user that, say, likes to tweet memes a lot. People that think this is an easy problem just don’t have the slightest idea of what it entails.
My point is it's an easy problem once you decide that you don't care about false positives, ie temporarily banning "legitimate" users whose behaviour is indistinguishable from that of bots.
Arguably, those "legitimate" users don't bring anything of value to the platform or society as a whole and should rightfully be shunned.
Most online forums or chats have automated rules whereby posting too fast or too often will get you slowed down or banned. Sometimes legitimate users get caught by the filters, no big deal. Twitter could make their existing rules much more stringent.
Once you add things like clickstream, email, ip address and user agent to the data used, it should be very easy to pick set with negligible false positives.
I actually am trending towards the view that once a social network reaches a certain size, all users must register their real identities with the network owner to maintain their accounts.
This information would not need to be stored, it would just be needed to verify an account as belonging to a real person.
The people complaining about censorship aren’t concerned about the removal of bots. They’re concerned about a pattern of behavior from twitter whereby they shadowban, remove, and editorialize viewpoints that don’t agree with the Valley’s ideology.
Alternatively, I wouldn't be surprised if a lot of the censorship complaints came from bots or people deploying bots that aren't happy they're being caught.
"The Valley's ideology" is to drag your feet, do as little as humanly possible, and to make vapid appeals to freedom of speech rather than do anything about rampant, massive-scale abuse of their services causing real and actual harm to people and society in general.
I know this isn't what you intended to say but it what your statement implies is that everyone who disagrees is uncaring, idiotic, or evil. It might do you well to spend some time finding people to respect on the other side and listen to why they advocate for things you find so distasteful.
I am trying to follow this supposed implication but I don't see the jump from "the valley" to "everyone who disagrees" and also from "dragging feet" and "making vapid appeals" to "idiotic and evil".
I will agree it's a bit of a stretch and I pattern matched to other arguments I have seen but "massive-scale abuse" and "real and actual harm" is language that presupposes a negative moral judgement about what is happening.
You think it is worth debating whether massive troll farms spreading misinformation and stoking hatred and division on a nation-state scale is good or bad?
None of these claims is real. “shadowbanning” doesnt exist, and that has been proven many times over. They also don’t remove any content that doesn’t infringe the TOS. This “valley ideology” nonsense is pure propaganda.
Lots of those studies pick a few criteria to label something "probably bot", with little verification that the mechanism is actually accurate. Which still makes for big headlines, but doesn't tell you if it would be effective and fair treatment if actually used to ban users.
(I'll add the link if I can find it, but a recent example pretty much boiled down to "actually, these accounts giving many repeated identical replies have totally normal timelines otherwise. And it turns out, many identical replies in short time shows someone knows about copy-and-paste, not that they're a bot")
Twitter's primary metric is "Monthly Active Users" which is driven by a) tweeting and b) engagement (aka activity) on those tweets.
It's not in their best interest to wipe out the bots at any large scale. Sure, the occasional cleanup or purge as a token effort or to get the most egregious stuff but remove them all? Nope.
I’ve worked at a global-scale platform for user-generated content doing anti-abuse and anti-spam tooling. In my experience those researchers do exist in-house and there was an imperative to remove abusive and spammy accounts.
The difference in my perception is that the presence of the bots and bad actors on Twitter that come to the attention of reporting like this increases engagement and views, and thus top line revenue.
This isn’t saying “bots count as views, so we get more ad dollars”. It’s that bots and bad actors promote topics and conversations that bring more real users to the platform, and increase the session duration for new and existing users.
I would imagine that Twitter sees spam bots and purveyors of illegal content as unwelcome and probably has an engineering team that dispatches those accounts quickly. But whether deliberately or unconsciously, they probably don’t apply the same rigor to accounts that break the TOS but manage to drive the top line up.
I’d love to hear from an engineer from Twitter who works in this space.
I see these "x number of bots found on Twitter" reports every so often. What are the most common methods used to determine if an account is a bot or not?
Adding an option to verify accounts by linking them to social security numbers or corporate registrations and marking them accordingly (without necessarily making the details visible publicly) could at least direct trust to accounts that can be traced to real entities.
This is an awful idea considering data breaches happen. Unless they figure out a way to store the SSN hashed so it cannot be reused, but SSN seems so fundamental that I’m not sure I’d trust any hashing algorithm for it. Especially considering the inputs are so limited.
The problem is that the US has no proper/well working ID systems. A large amount of social security numbers had been compromised through leakage and are hardly reliably usable for identification anymore.
Through there are many countries in which this can be indeed done reliably sand effectively.
Through the main problem is still this networks are global, so bots will just register with origins where IDs can be easily faked/stolen etc.
But it might still help a lot for local discussions if you would make it clearly visible if a person had a no or a non local real person identification.
If such a system is done cleverly it could also help law enforcement while it still upholds privacy. Through it's will always be someway prone to abuse by police and similar if not done perfectly transparent but you can be sure that certain lobbies will invest insane amounts of money into making it non transparent over time. Which makes such systems potential dangerous to have.
> Adding an option to verify accounts by linking them to social security
Not sure if you’re joking, but this would be a completely awful privacy situation. Why would you ever give your social security number to a social media site?
Kidding, but there's a real point. 40% of tweets are from bots? OK, how many "studies" are from bots? (I believe that's in the neighborhood of Drybones' point.) But how many comments responding to the studies are from bots?
Way back in the day, I remember reading the Cluetrain Manifesto. It said that corporate-speak sounded literally inhuman. Well, the bots have gotten better over the years. But also, it seems to me that the humans have gotten worse. They are less able to say "Wait, that doesn't make sense." They are even less able to say "Wait, that doesn't sound like the way humans talk." And that's bad. As the bots get better, the humans need to be developing better filters, not regressing.
I’m not sure why Twitter bots matter, and I assume the only point is to discredit the argument of people against lockdowns.
Real people have also been out protesting the lockdowns, in the real world, not on Twitter. People were legitimately upset, rightfully so, when protestors were showing up with guns in Michigan.
It matters if it means that much less people then believed do protest. I don't know if it does mean it, it could, but it also could not. It also matters because it means some people try to manipulate what is perceived as the "Twitter public" opinion.
That's why one never counts "twitter public", but real protestors. A small silver lining to coronavirus is that crowd size estimation is much easier when the crowd is on a 2m lattice.
I use to hear a lot they in customer service one angry customer might represent only one person, but more than one on an issue could possibly point to an exponentially larger dissatisfied group. People who complain versus people who do not, people who call versus email, etc. I do not know if this is still accurate.
Somewhere I'd read a book on propaganda by a fellow who produced it in ww2 and finished his introduction with biblical examples of the main techniques.
People are influenced by the words and appearance of others... If people think that there are many other "supporters" of something, they'll be far more likely to support said thing. You are as influenced by this as anyone.
Personal attacks and name-calling will get you banned on HN. If you wouldn't mind reviewing https://news.ycombinator.com/newsguidelines.html and sticking to the rules when posting here, we'd be grateful.
I think we have to except that there are bots everywhere, here, Wikipedia, read it, Twitter, etc. It doesn’t appear to be slowing down but instead increasing in growing in magnitude, if this article is to be believed it doesn’t appear to be slowing down but instead increasing and growing in magnitude, if this article is to be believed.
Bunch of threads on twitter from expert researchers on how this study is bogus. The main point is that there is zero paper or even a blog post that can be referenced to see if this statistic is true.
My takeaway is that "news" from commercial outlets like MarketWatch (the publisher here) largely is not fact-based. Compare to reports in February about waves of anti-Asia discrimination, correspondeding to a major and ill-advised Democratic information initiative.
I don't know if Twitter has bots or how they tilt. What I do know is post-coronavirus commerical "news" outlets have lost my trust, as has much of the political establishment, red or blue.
For example, MarketWatch has been pushing articles for months about discrimination relative to public health measures to generate a narrative critical of the U.S. travel ban earlier this year affecting flights to and from China.[1] Closer to home, you can look to the steady diet of fear-based public health reporting that the "news" outlet finally has explicitly disclaimed as non-factual (but always was defective) -- "Due to the rapidly changing nature of the coronavirus public health crisis, some of the recommendations included in this article may eventually become out-of-date."[2]. It's ok to be wrong, but the "news" consistently has been "wrong" in a manner that has generated support for particular political policies, even particlar drugs have become political lightning rods whose efficacy is not being investigated or reported in a fair and reliable manner.[3]
One can explain this, and explain that, but when the sum result is being locked inside for months on end, unemployed, blocked from major life events with a media wrongly cheeleading those policies with wrong and unsupported information, this leads to a loss of trust. I'm not saying the media outlets are on a lark; I'm saying they haven't done their jobs of reporting facts outside the filter of propaganda and political bias which has infected so much policy dialog.
> Closer to home, you can look to the steady diet of fear-based public health reporting that the "news" outlet finally has explicitly disclaimed as non-factual (but always was defective) -- "Due to the rapidly changing nature of the coronavirus public health crisis, some of the recommendations included in this article may eventually become out-of-date."[2].
Well, let's see.
> steady diet of fear-based public health reporting
There's a pandemic on with a nontrivial death rate. A little fear's deserved.
> that the "news" outlet
I believe these are called scare quotes.
> finally has explicitly disclaimed as non-factual
where is this disclaim?
> (but always was defective)
You have certainty on your side apparently.
Now the quote about stuff getting out of date, well, what do you expect? Speak to god if you want utter certainty, not scientists. Seriously, what makes you think eternally correct information drops fully formed into anyone's lap.
If you want to live in a world without science, aka the middle ages, I urge you to do so, so those of us left behind with the internet, antibiotics, human rights, a stable civil society, non-agricultural jobs, and education, can suffer alone.
This quote from the CMU article lines up with the headline on this HN discussion thread (200M, 62%, bots, COVID-19):
> CMU researchers since January have collected more than 200 million tweets discussing coronavirus or COVID-19. Of the top 50 influential retweeters, 82% are bots, they found. Of the top 1,000 retweeters, 62% are bots.
Fascinating article. Thank you for taking the time to find it.
Almost all "research" I've seen in this domain is politicized bullshit. I assume even if it isn't, the ones that get air do not get it because of merit; but because it suits someone's political axe-grinding.
Why the ranting shouts of 'bullshit' and claims of 'political axe-grinding'? Posts like this, that rarely have any facts with them, just suggest someone's angry at something they don't like. They don't tell me anything except someone's temperature is rising.
The "ranting shouts" are justified because, like the whole "weapons of mass destruction" pile of bullshit, touted by mostly the same goddamned weasels -there is zero evidence for their assertions, and ample evidence that they're lying: again.
If you google my name, unlike your "throwaway" anon name making assertions, irony of ironies, you will see I am a real human being, fully qualified to check into data science type assertions like this (and in fact; I have). Which is a lot more than "you" can say.
The idea that half of Twitter's accounts are bots is not new: https://gizmodo.com/how-many-social-media-users-are-real-peo... (written in 2018)
Another older article puts the issue into a larger social context and makes some sloppy estimates:
"Since Twitter’s yearly revenue is between 2 and 3 billion dollars — that is as much as 50 dollars per potential customer on average and that doesn’t count what people are spending to purchase fake followers and bots. If only English speaking customers are targeted by Twitter ads, that means 150 dollars is being spent to grab the attention of each customer."
https://kirstenhacker.wordpress.com/2019/08/05/the-fake-econ...
Personally, I watched a guy who was promoting his crappy, plagiarized book go from 25k followers to 90k followers over the course of a week. They were all clearly fake since he rarely got more than a single like on everything he Tweeted.
I'd like to know what "bot" means. Is it the new word for internet troll or do these Twitter accounts really have better AI than Microsoft or Facebook.
Bots can be used for many purposes. Many (maybe most?) are constructive. There have been a number of DIY tutorial for beginners about how to use Arduino or Raspberry Pi to let plants tweet when they need water. Here’s one (old) example: https://www.wired.com/2008/02/houseplants-wil/.
Not surprising at all. From my experience with Twitter, I would say 45-50% of all tweets come from bots.
I feel like there are 4 types of Twitter users: 1) bots, 2) influencers that heavily use automation to the point where they are almost the same as a bot, 3) famous/popular that are somewhat to mostly real, but have someone else managing their account for them, 4) real people who think 1-3 are all real that gained their followers organically.
Probably most users fall on categories 1 and 4 above, although most of Twitter traffic is probably generated by tweets of 2 and 3.
You mean both 80% of the users are bots and 80% of the tweets are from bots? I would find that fairly surprising that bots and users work out to be the same proportion in both of those.
Sure, there are indeed bots on twitter, mostly for advertisements, but I kinda doubt that there are bots operated by states. That being said, even if such bots existed, what would be the point? The only effect seems to be that you have people from both camps accusing each other of being bots. In addition to quote the user repolfx from https://news.ycombinator.com/item?id=20776752
> Namely, their rationale was something like this: the world is dominated by people whose opinions are fundamentally shaped by what others are saying, and moreover, by how frequently other people seem to be saying it. They don't really think about things or rely on their own experience: they're just mimics.
> In such a world being able to auto-generate fake messages of support for some political position or another at scale would give you immense power, because the population would automatically swing behind you based merely on the perception that everyone was swinging behind you.
> So that's their fear. But is it realistic?
> Well, this is where we get into the polarisation. "You aren't smart enough to have an opinion" is the sort of viewpoint that leads to an elitist vs populist conflict. There have been reams of analysis about this and it's not really AI specific - e.g. did people vote for Brexit because of Twitter, or because of things they saw in the press, or did they vote based on their own experiences, or what their close friends/family thought, or what mix of "all of the above" is the truest mix?"
169 comments
[ 3.0 ms ] story [ 244 ms ] threadThe issue they push with the bots may not be, and likely isn't, their actual position on the issue.
Too bad we can't have a Geneva Convention on botnets or something. (I would write "couldn't", but don't believe it's technically feasible)
[1] https://en.wikipedia.org/wiki/Foundations_of_Geopolitics#Con...
I think Twitter needs to add an option which marks an account as a bot or non-human account so that people can gauge instantly what the account's real motivations are. Most actual bots will ignore that setting and pretend to be human however, so the responsibility is on Twitter to weed out automated accounts that are obvious attempts to game the platform.
It's not unlikely the "So?" is meant as an "alright then, can we now have the answer to ...". I've made that exact mistake.
https://xkcd.com/2305/
As for twitter bots, I have one running by ifttt, another by script, and three more in planning. I welcome the human/bot accounts being made clear.
https://www.usatoday.com/story/news/nation/2020/05/29/corona...
That said, with reading this, if you are thinking of skipping it. Does somewhat confirm that most of the not posts are on conspiracy theories. Which, I guess is not that surprising. Even if the conspiracies are.
I often wonder when we are going to see more left wing reactions to this sort of behavior. Seems the right, based on my limited information, seems to be more aggressive about these disinformation campaigns.
Also it stands to reason that this particular misinformation campaign is being sponsored by a foreign state. I think that this would be an effective weapon to use against the United States. Sort of like biological warfare by proxy of disinformation.
Perhaps we should all step back and evaluate whether our internal thought processes are being shaped by bad actors via an appeal to our innate tribalism? Notice how easily you fall into the script: 1) it’s the other team only 2) my team should do something to fight back 3) maybe it’s sponsored by a foreign state.
If you primarily think one side is crafting ridiculously bad content and it’s therefore a misinformation campaign from a foreign state, I think it’s wise to consider how much of that is personal bias and if you’re perhaps “giving a pass” to stupid content that at least aligns with your views.
“Well, they meant well, so it’s probably just a right-minded American who made a simple off-by-factor-of-a-million arithmetic error and no one spent 50ms doing a basic sanity check. No one’s perfect.”
Or “Look, maybe I think Obama did an Ok job, but I saw all that content about he wasn’t born here, and there has to be something to it if so many people are talking about it. The government hides things; the truth is out there, ya know? Oh, and Epstein didn’t kill himself.”
Again, I said from my "limited view", as it's impossible to have all the information. It seems TO ME that there is quite a bit more targeted misinformation coming from the right.
I would love to be corrected. However, I don't think saying "I got a Facebook feed with lots of stuff stuff coming out of it", and "you're stupid because of xyz I assume about you for no reason whatsoever" qualifies in this regard.
I’m sorry for the portion of that where the fault is mine.
However, I am concerned about a disinformation campaign which could, theoretically, be sponsored by a foreign state to get people to ignore the very real threat of coronations.
I am also concerned about foreign influence, but I also observe that we meddle in other countries’ affairs as well, so it’s just a fact of life and social media changes it but didn’t start it. I’m more concerned with a population who isn’t thinking critically and with the right balance of short and long term (and is therefore especially susceptible to influence), but whether the influence is foreign or domestic doesn’t matter as much to me as the effect of the influence.
Is this even disputed?
https://news.ycombinator.com/newsguidelines.html
It would be much better to show useful statistics to convince people that bots are kept to a minimum.
I think even if you were to present statistics of some kind, some people might not consider it enough and ask for some different contrived set, on and on, forever.
If you think something's really wrong, you email the mods about it. Otherwise you get in this exact meta-discussion land about what is astroturfing, what are accusations, and so on. And then you get threads that distract from meaningful conversations, on and on, forever.
Maybe we could all just try to discuss the article and related comments by refuting/agreeing with their central points. It would be less boring than seeing the same metadiscussion and flailing forever.
(In that vein, I won't be replying to this comment further)
edit: for those left wanting, take a look at the words of 'dang: https://hn.algolia.com/?query=by:dang%20astroturfing&sort=by...
If it happens a lot and without a given reason it certainly does, but it isn't such a black and white issue that it has to be stopped at all costs. When someone points out very suspicious patterns over multiple names, comments, similar threads that are more likely to have astroturfing or bots etc. and mods pop in to say not to talk about astroturfing and bots because it doesn't happen, that's not helping keep the site to a higher standard.
I don't doubt it is a priority and people are working on it, and I don't think the reply here that has no examples is worthwhile, but an all or nothing attitude doesn't help, because it is the reality of the internet.
It's also not so much about trying not to make "the site look bad", it's more that it generally leads to baseless witchhunts and flame wars.
HN has an advantage over Twitter and other social networks: posts are supposed to be insightful instead of merely pictures of cats or people doing funny tricks. As a result we can criticize posts based on their intrinsic values instead of having to go meta and wonder about the intentions of the person posting them.
If a bot manages to posts dozens of interesting stories that make it up to the top of the front page every day I say we keep it...
Also, from time to time I see accusations that anti-CCP comments/articles are flagged to death by bots, but I've never witnessed this first-hand.
If you then downvote those posters you disagree with, it creates a very suspicious voting pattern on their end.
Could you distinguish your own behavior from that of a bot (or a bot-assisted human) if you didn't know it was just you in front of your screen?
(I’m not even sure I’d consider something that was itself read-only in nature to be a “bot” but I suppose it technically is.)
It's also worth noting that some of the worst stories that have resulted in the most division have been ideologically driven. I recall a libertarian think tank that had a number of accounts that apparently existed purely to make submissions for those sites. I'm sure there are other groups that are at it too, I just found that one to be particularly blatant.
Regardless of your political views, the submission of such stories seems to inevitably result in a flame-war that seems only to serve to degrade this community's technologically-orientated mutual respect.
It's even crazier that Twitter can't do that research themselves and remove those bots.
Obviously the big debate is censorship. It feels like there could be a tech solution to this, where conversations aren't shaped by volume, but I guess that is what Twitter is anyway.
It’s a lot harder problem then it might sound on the surface. And realize if you find bots 100%, they will change for round 2 and the game starts over.
Just put in a super basic unlock process that is costly to game for "non-real" people
I think your "just" is hand-waving away a lot of complexity. A person deployed these bots, and if a bot account gets put into an arbitration process, that person can go through the process themselves to pass the arbitration, then set the bot running again.
That is not scalable, which is exactly the point.
(Assuming Twitter was actually incentivized to be anti-bot/bot-like behavior.)
And it wouldn't even stop the bots, because Twitter isn't limited to Americans, so bot makers could easily send fake official IDs from someplace Twitter can't check. Or send stolen offical IDs from other data breaches.
You can generate realistc photos of IDs from countries that Twitter cannot positively verify those bc in USA there are ways to check authentity of your document. But IDs from Russia or Ukraine - no way.
That would work only against casual bots. The big nation-sponsored botnets won't have issues with providing genuine government issued identities for bots. Which kind of defeats the whole purpose and wastes a lot of resources twitter would like to use elsewhere.
Whenever I sign up at Twitter (entirely because some companies do support there) the account is flagged within a few minutes for "suspicious activity", such as making a coffee and coming back to the PC or searching for the company I want to contact.
Think about an account that is active 24/7 every day all year, or one that fires off multiple actions per second for long periods of time.
Beyond that, having a limited vocabulary, repetitive phrases, consistently accessing the same thing every few minutes, etc. are all low hanging fruit.
Arguably, those "legitimate" users don't bring anything of value to the platform or society as a whole and should rightfully be shunned.
Most online forums or chats have automated rules whereby posting too fast or too often will get you slowed down or banned. Sometimes legitimate users get caught by the filters, no big deal. Twitter could make their existing rules much more stringent.
They also have so many phone number and email they can use that to do additional checks.
They you can use the network effect: bots usually have something in commons: topics, likes, ip, followers...
If you have millions at your disposal, something can be done.
They are not, so I'm more encline to assume there is benefit for them to avoid doing so.
This information would not need to be stored, it would just be needed to verify an account as belonging to a real person.
a) it was a non profit and
b) people didn’t cry “censorship!” every time they do anything to curb abuse and
c) it would be ok to deactivate a % of real users by accident (these researchers can’t tell who’s a bot vs who’s an actual user _for sure_)
It’s not an easy problem.
Dude, you can’t even form a grammatically correct sentence.
(I'll add the link if I can find it, but a recent example pretty much boiled down to "actually, these accounts giving many repeated identical replies have totally normal timelines otherwise. And it turns out, many identical replies in short time shows someone knows about copy-and-paste, not that they're a bot")
It's not in their best interest to wipe out the bots at any large scale. Sure, the occasional cleanup or purge as a token effort or to get the most egregious stuff but remove them all? Nope.
The difference in my perception is that the presence of the bots and bad actors on Twitter that come to the attention of reporting like this increases engagement and views, and thus top line revenue.
This isn’t saying “bots count as views, so we get more ad dollars”. It’s that bots and bad actors promote topics and conversations that bring more real users to the platform, and increase the session duration for new and existing users.
I would imagine that Twitter sees spam bots and purveyors of illegal content as unwelcome and probably has an engineering team that dispatches those accounts quickly. But whether deliberately or unconsciously, they probably don’t apply the same rigor to accounts that break the TOS but manage to drive the top line up.
I’d love to hear from an engineer from Twitter who works in this space.
Through there are many countries in which this can be indeed done reliably sand effectively.
Through the main problem is still this networks are global, so bots will just register with origins where IDs can be easily faked/stolen etc.
But it might still help a lot for local discussions if you would make it clearly visible if a person had a no or a non local real person identification.
If such a system is done cleverly it could also help law enforcement while it still upholds privacy. Through it's will always be someway prone to abuse by police and similar if not done perfectly transparent but you can be sure that certain lobbies will invest insane amounts of money into making it non transparent over time. Which makes such systems potential dangerous to have.
Not sure if you’re joking, but this would be a completely awful privacy situation. Why would you ever give your social security number to a social media site?
Kidding, but there's a real point. 40% of tweets are from bots? OK, how many "studies" are from bots? (I believe that's in the neighborhood of Drybones' point.) But how many comments responding to the studies are from bots?
Way back in the day, I remember reading the Cluetrain Manifesto. It said that corporate-speak sounded literally inhuman. Well, the bots have gotten better over the years. But also, it seems to me that the humans have gotten worse. They are less able to say "Wait, that doesn't make sense." They are even less able to say "Wait, that doesn't sound like the way humans talk." And that's bad. As the bots get better, the humans need to be developing better filters, not regressing.
Real people have also been out protesting the lockdowns, in the real world, not on Twitter. People were legitimately upset, rightfully so, when protestors were showing up with guns in Michigan.
Edit: not very useful these days, but still of possible interest: http://www.gkstill.com/Support/crowd-density/100sm/Density1....
Somewhere I'd read a book on propaganda by a fellow who produced it in ww2 and finished his introduction with biblical examples of the main techniques.
https://news.ycombinator.com/newsguidelines.html
* https://twitter.com/ngleicher/status/1264614994475315200
* https://twitter.com/benimmo/status/1265329734705197056
* https://twitter.com/alexstamos/status/1264643202293751808
* https://twitter.com/3r1nG/status/1264567090742206464
(edit to add more examples of counterclaims to this "research")
I don't know if Twitter has bots or how they tilt. What I do know is post-coronavirus commerical "news" outlets have lost my trust, as has much of the political establishment, red or blue.
Can you please share examples of specific news reports that we’re non-factual?
One can explain this, and explain that, but when the sum result is being locked inside for months on end, unemployed, blocked from major life events with a media wrongly cheeleading those policies with wrong and unsupported information, this leads to a loss of trust. I'm not saying the media outlets are on a lark; I'm saying they haven't done their jobs of reporting facts outside the filter of propaganda and political bias which has infected so much policy dialog.
[1] https://www.marketwatch.com/story/no-chinese-allowed-racism-...
[2] https://www.boston.com/culture/lifestyle/2020/04/16/is-it-sa...
[3] https://www.theguardian.com/science/2020/may/28/questions-ra...
Well, let's see.
> steady diet of fear-based public health reporting
There's a pandemic on with a nontrivial death rate. A little fear's deserved.
> that the "news" outlet
I believe these are called scare quotes.
> finally has explicitly disclaimed as non-factual
where is this disclaim?
> (but always was defective)
You have certainty on your side apparently.
Now the quote about stuff getting out of date, well, what do you expect? Speak to god if you want utter certainty, not scientists. Seriously, what makes you think eternally correct information drops fully formed into anyone's lap.
If you want to live in a world without science, aka the middle ages, I urge you to do so, so those of us left behind with the internet, antibiotics, human rights, a stable civil society, non-agricultural jobs, and education, can suffer alone.
https://www.cs.cmu.edu/news/nearly-half-twitter-accounts-dis...
I imagine they're going to publish the research, but it doesn't look like they have yet.
...though some of their methods are described.
> CMU researchers since January have collected more than 200 million tweets discussing coronavirus or COVID-19. Of the top 50 influential retweeters, 82% are bots, they found. Of the top 1,000 retweeters, 62% are bots.
Fascinating article. Thank you for taking the time to find it.
If you google my name, unlike your "throwaway" anon name making assertions, irony of ironies, you will see I am a real human being, fully qualified to check into data science type assertions like this (and in fact; I have). Which is a lot more than "you" can say.
It's all subjective anger and whatnot. What am I or anyone supposed to take away from that?
> ample evidence that they're lying: again
Ok, what evidence. And don't tell me to google it or I can find it if I want to. You claim, you cite.
> You claim, you cite.
If a claim cannot be verified, it is not a claim per se: it's an opinion.
* 50% of positive posts for candidate x are bots!
* Most people who support X online are bots!
Easy and cheap, and the message carries well.
That condition is the key. Are you sure they can do that? It’s very tough to do with certainty.
I feel like there are 4 types of Twitter users: 1) bots, 2) influencers that heavily use automation to the point where they are almost the same as a bot, 3) famous/popular that are somewhat to mostly real, but have someone else managing their account for them, 4) real people who think 1-3 are all real that gained their followers organically.
Probably most users fall on categories 1 and 4 above, although most of Twitter traffic is probably generated by tweets of 2 and 3.
It almost feels like playing a video game where most other characters are NPCs pretending to be real users.
> Namely, their rationale was something like this: the world is dominated by people whose opinions are fundamentally shaped by what others are saying, and moreover, by how frequently other people seem to be saying it. They don't really think about things or rely on their own experience: they're just mimics.
> In such a world being able to auto-generate fake messages of support for some political position or another at scale would give you immense power, because the population would automatically swing behind you based merely on the perception that everyone was swinging behind you.
> So that's their fear. But is it realistic?
> Well, this is where we get into the polarisation. "You aren't smart enough to have an opinion" is the sort of viewpoint that leads to an elitist vs populist conflict. There have been reams of analysis about this and it's not really AI specific - e.g. did people vote for Brexit because of Twitter, or because of things they saw in the press, or did they vote based on their own experiences, or what their close friends/family thought, or what mix of "all of the above" is the truest mix?"