Twitter Is Crawling with Bots and Lacks Incentive to Expel Them
bloomberg.com
bloomberg.com
The definition of a bot could be problematic though. If a bot is merely automated posting, then virtually every company, news org, website, and anything else that uses automated posts is a "bot", and are those bad? I don't think so, those are useful. The problematic bots are basically automated trolls.
Troll bots increase "engagement" metrics by quite a bit, this is obvious to any person if you look at the automated bot replies to virtually any post from a news agency or anyone with a large twitter following. They also jump onto every hash tag or breaking news event spreading garbage. You'll find dozens of bots making absurd statements followed by bots responding to those bots, and clueless people arguing with the bots. Combine that with the endless seas of human trolls that also jump into everything, and those interactions probably make up a huge amount of usage of Twitter.
Twitter remains useful for scanning headlines and breaking news, but outside of that the service is a complete mess.
It has a Klout score. It has more followers than me.
[edit: spelling]
If it becomes self-aware, we may have a problem.
I've seen this movie before somewhere...
Klout still exists, apparently. I don't see many people referencing it anymore though.
The bad ones though ruin the experience for everyone (and defraud advertisers).
I've previously attempted to classify bots somewhat:
https://medium.com/@iTrendTV/dumb-bots-have-ruined-twitter-s...
Many users use some automation and that’s a good thing, e.g. posting blog posts when published, bots to repost top #shithnsays for those who want to follow it, that amusing horse books bot etc.
More problematic than bots to me are the many many ways the twitter platform simply rewards the wrong thing, from verified nazis to crowds of trolls and spamming hashtags.
Also most of the problems on twitter are caused by humans, not bots. Humans can be cruel, harass, bombard victims with messages, threaten them with death, rape etc (this is regularly happening right now to women on twitter), and those are the more problematic interactions. Yes sometimes spambots liking posts or replying to posts are annoying, or the ones that DM you are pretty bad, but they're not as annoying as humans at their worst.
Yes. Make bots legitimate, with verified bot status for ones that are truthful and helpful to the end users.
I use Twitter daily, but mostly as a news feed, or with people I already know in real life. There is no point in engaging with a random person on Twitter, because it's quite often a bot or someone that wants a fight.
Except if you see that for example the BBC's World News account is a bot, that makes sense because the BBC isn't a person, it's an organization.
The problem is bots masquerading as individual people.
- A reason not to pay to advertise on twitter
- A way to make the platform look more active than it is
- A distraction to users looking for conversations
Is it a place for people to chat with friends?
Is it a platform for free speech or are there limits to that speech?
Is it a platform for brands/people to broadcast to a broad set of followers?
Is it a source of news and a place to talk about politics or what is going on?
It can't be all of those things.
It seems it was not designed with any purpose or philosophy in mind.
It is not good for reading text. Not good for reading conversations. Not good for looking at images.
I do admit that they got the scroll until your finger is sore thing down.
"According to a recently published report, Facebook says they reach 1.5 million Swedes between the ages of 15 and 24. The problem here is that Sweden only has 1.2 million of ’em"
"Facebook’s Ads Manager says that the website is capable of reaching 41 million Americans between the ages of 18 and 24. The problem is there are only 31 million Americans of that age." [1]
Nothing ever happens, other than maybe some grumbling from paying advertisers.
[1]https://mumbrella.com.au/will-facebook-ever-stop-bullshittin...
(Via Matt Levine, tongue firmly in cheek)
http://www.thevab.com/wp-content/uploads/2017/09/Facebooks-R...
trying to police them off would be a mistake by facebook.
I'm sure they do, but I doubt 30% of them do. That's what you'd need to balloon from 30 million to 40 million. The 30% also assumes that EVERY 18-24 year old American has a Facebook account in the first place.
I've seen people create new accounts every time they lose their password. I also had a friend who'd create a new account every time she had a personality change of some sort (new hair? new profile!). Another friend owns a couple businesses and has accounts for each one (Pages are completely lost on him...). My aunt has entered her birthday incorrectly so she's supposedly 19 years old right now. A neighbor has a "family account", which is really just the wife posting pictures of kids.
Don't take "technically inept" to mean "my abilities, just a bit lesser". Take it to mean "no ability to reason about software applications whatsoever".
Edits: Correcting math. Thanks, kbenson.
Most like the .01% have hundreds to thousands of accounts. What does that imply?
At some point should there be regulation on what terms like MAU means?
Right now, investors researching Facebook stock have to be aware that published metrics include botspam. Suppose an authority imposed a regulation to force Facebook to publish "bot free" metrics.
Then what will happen? Legal will impose a demand on engineers, and the engineers will scratch their heads and do what it takes to satisfy legal, and legal will scratch their heads at the result and write up whatever report it takes to satisfy the regulator.
In the end, the relationship of the published number to reality will have changed in some difficult to predict, probably undocumented, way. Will investors then really have an easier time doing their research?
OTOH, it's pretty tempting to use numbers already pre padded for you.
There may be real people behind them, but they are "employed" by various agencies as proxies for RT/likes. Some do it manually, others employ automation.
Also, many of the more prominent accounts are almost fully scripted - these people aren't really "there". Here's an example of a Salesforce exec (Verified account, btw) employing "automation":
https://twitter.com/iTrendTV/status/738375663229505536
We've done some twitter fraud analysis in the past, and it's terrifying.
https://twitter.com/Prudential/status/918475795479351296
Click on the Likes below the tweet (will show you list of accounts engaged), and tell me this is real human engagement.
You will see accounts like this:
https://twitter.com/Electriflyy
Twitter won't consider this a bot, yet this "person" already liked 192,000 tweets:
https://twitter.com/Electriflyy
Here's another account who engaged with this promoted tweet, for which Prudential is getting billed:
Again, look at the number of tweets this "person" liked (72,000).
Other "engagements" come from:
https://twitter.com/RobertT70111579 https://twitter.com/TeghanJustin https://twitter.com/brandon_gorsuch ... Notice the pattern?
Prudential is getting ripped off right in front of us.
This is happening now across all brands, at massive scale.
"Engagements are new followers from Promoted Accounts and/or clicks, retweets or likes on Promoted Tweets. You will never be charged more than your maximum bid and you usually pay less."
https://business.twitter.com/en/help/overview/twitter-ads-gl...
The bulk of all engagements now are fake/automated.
P.S. as a bot, you don't need to "see" a tweet to engage.
The SEC protects investors, not advertisers. When non-securities fraud gets companies in trouble with the SEC, it's because said companies failed to disclose the fraud or the risk of the fraud to their investors.
Facebook discloses their "advertising revenue could also be adversely affected by a number of other factors, including...the availability, accuracy, and utility of analytics and measurement solutions offered by us or third parties that demonstrate the value of our ads to marketers, or our ability to further improve such tools; adverse legal developments relating to advertising, including legislative and regulatory developments and developments in litigation; decisions by marketers to reduce their advertising as a result of adverse media reports or other negative publicity involving us, our advertising metrics, content on our products, developers with mobile and web applications that are integrated with our products, or other companies in our industry..." [1]
[1] https://www.sec.gov/Archives/edgar/data/1326801/000132680117...
Disclaimer: I am not a lawyer. This is not legal advice. Don't commit fraud.
Are they reporting user numbers or account #s? Is it ok if they generate 1billion accoints themselves? Why/why not? What's different?
Impressions are based on "eyeballs" on ads. Machines look at ads. Customers can project ad blocker %.
The intentions are clear - what will happen is not. The problem is you cannot prove malicious inaction.
[0] https://newrepublic.com/article/125952/gop-party-fear
[1] http://thehill.com/homenews/senate/352768-gop-senator-russia...
It's not a GOP- or Democrat-specific problem.
Plus, it's not like there's only "right-leaning" bots on Twitter. Back during the election there were a couple times I tweeted something negative about DJT and my tweet was instantly retweeted and liked thousands of times by blatantly "left-leaning" bots. They hit up both sides of the aisle to stir up drama. It's practically their stated goal.
When two guys fight they both get a black eye. It is the third guy who got them to fight that gains the advantage over them.
This is correct until you ask "how much" does one benefit from divisive speech.
There is one nationalist branch of the Owner's Party of America, and the other is Democrat.
[0] http://thehill.com/policy/technology/351557-russian-actors-m...
Also, it requires a rather curious definition of "bots" to include actual people manually doing stuff like this. One that's political rather than technical.
So either they're inflating numbers and lying to advertisers... or they're capping numbers and lying to make the "deplorable" subs look less popular.
The problem isn't just bots, but fraud in general.
I went into trouble of reporting each of the fake accounts manually (the UI for this is far from convenient). After some time I checked the bot accounts again. None of them was deleted, our original tweet still had a large re-tweets count, which basically misleads the real Tweeter users, making the tweet look more interesting that it really was.
From this incident I no longer look at re-tweets as any reliable metric of popularity or quality of tweeted content.
Unfortunately, bots don't just inflate user numbers for them, they create ad revenue.
Additionally, the majority of self-proclaimed "influencers" rely almost entirely on automation for publishing (i.e. they are not really "there" on the platform), and bots for amplification.
If you dig deeper, there's a number of "certified" Twitter partners there, that provide social media analytics and management platforms. They pay Twitter for data access, but they employ 3rd parties to provide "amplification" for them, via bot farms - creating the illusion of effectiveness for the subscriber.
The rate of real, authentic human discovery and engagement on Twitter right now is incredibly low.
How? At the end of the day, a human sees the ad and so generates ad revenue for Twitter. A bot a) does not see the ad tweets in his normal timeline feed and b) does not feed back impression tracking to Twitter.
If you promote a tweet or account on Twitter, you will get billed for Engagements (RTs, likes, follows).
https://business.twitter.com/en/advertising/campaign-types/i...
http://adage.com/article/digital/brands-worry-twitter-undere...
Source?
I thought a year or two ago that Twitter was avoiding cleaning house and removing the obviously fake accounts because they were simultaneously trying to show growth on the platform and if anything removing these bots would hurt their growth metrics.
Although they technically ban botting, and make cursory efforts to stop it, it's not really stopped.
Here's a good article on Insta botting, which goes info further detail: https://petapixel.com/2017/04/06/spent-two-years-botting-ins...
If someone has a legit reason to fear registering a bot they run, then it's probably for "not-good" reasons. And in my estimation "not-good" could be as innocuous as those follower-fishing bots.
The "up-to-no-good" bots do no such thing, of course.
Since I've never personally tried to solve any of the problems on that imaginary list, it's tough for me to discern between:
1. It's hard
2. It's possible, but has bad side-effects (like censorship)
3. No incentive (ultimately will not help their stock price)
Someone on Wall Street is. And as long as there is a sucker to be sold to, the charade will continue.
Ex.
- make a bot network of 100 bots.
- identify some accounts, split them into control and experimental groups
- choose a topic, such as "NFL"
- every time someone in experimental group tweets something about the NFL, like or retweet it a dozen times from a random slice of your bots
- do this for a few weeks
- at end of few weeks, does experimental group tweet about NFL more frequently than control? Has the sentiment within their tweets gotten any more extreme?
@newsyc250 @newsyc100 @newsyc50 @newsyc20
Why should these bot accounts be expelled? I just love the way some bot accounts are so useful. In fact, following the above mentioned bot accounts, gives you just enough number of Hacker News posts that you want to see, based on how popular they are.
Similarly, there can be many bad bot accounts, but differentiating those from good bot accounts may not be an easy task, and bound to fail with false negatives and false positives.
But from a legal point of view, is Twitter obliged to find and ban such bots? I don't think so. Marking them as bots would be a good idea though.
I agree. I like how Telegram does bots. They have first class support for bots, and the separation between real user and bot is extremely clear. You have to specifically create a bot account, and bots have certain restrictions that users don't have (e.g. they cannot initiate a conversation themselves).
Of course, that assumes the bot is set up as intended. One could masquerade a bot as a real user if they were clever enough and had a spare phone number, presumably. I've not tried it, but I'm going to assume it's possible, since you can very well write your own Telegram client.
If twitter can fix that, it’ll be awesome.
> "Violent threats (direct or indirect): You may not make threats of violence or promote violence, including threatening or promoting terrorism."
And that's just getting started.
Has there been ANY statement from Twitter leadership on why they permit Trump to continue with this behavior? I have my own cynical answer, but I'm curious if they've gone on record for what is such a blatant abuse of their ToS it is unconscionable that they let it continue.
> Surely threatening nuclear war violates their "Abusive Behavior" rules.
Even if you were to try to live by the letter of the law, you'd have a hard time really getting this to stick, I imagine. If this is the tweet you're referring to:
"Just heard Foreign Minister of North Korea speak at U.N. If he echoes thoughts of Little Rocket Man, they won't be around much longer!"
Then I think you'd have a hard time construing this as "terrorism", any moreso than half of Twitter's population has said (sports teams getting "killed", public figures "not being around anymore" (i.e. probably meaning "in office")).
(Not to mention that that the main reason the demands to ban Trump have restarted is because Twitter briefly suspended someone for posting another person's private phone number to their 800,000 Twitter followers, and she and everyone else is using this to justify why Twitter should have let her get away with it.)
When you take a stand for something - you might break some bridges - but you'll also be building bridges.
It would be pretty interesting to see what the response would be, because Trump would no longer have an easy, low-effort medium to get his message out. Would he just turn to a right-wing social network like Gab? Would he just use traditional media instead?
Extremely unlikely.
http://huffingtonpost.ca/2017/08/18/donald-trump-worth-2-bil...
This clause is pretty clearly about illegal violence. Part of the President's job can involve both threats and actual violence. I think this president is a dangerous loon. But I don't think Twitter should a priori ban government accounts from talking about, say, a declared war.
That kind of talk has no place on Twitter based on Twitters own rules that say no threats are allowed.
The words are an expression of the ideas in people's heads, and are written with a particular context in mind. That context is fighting the normal kinds of platform abuse they see. The policy was definitely not meant to encompass the typical actions of governments and heads of state. Nobody was even thinking about that at the time they were written.
If they were to use that to ban Trump, it would be pretty obvious rules lawyering [1]. Which many people certainly feel is justified, and I understand why. But from Twitter's perspective, rules lawyering, no matter how much people feel it's justified, undermines trust. So if they want to get rid of Trump, they'd be better off writing a specific policy that Trump (and others) are clearly violating, not bending something else to fit.
https://www.avclub.com/twitter-releases-statement-confirming...
It would certainly gain a lot of positive attention and give huge props to Twitter IMHO too.
From your tribe, and it would generate a lot of negative attention and give huge condemnation to Twitter IMHO too from the other tribe.
> Because they're acknowledging and disallowing abusive behaviour?
They might have a bit more of a leg to stand on if their abuse council had a wider political spread or they actually policed threats that are illegal in a consistent manner. They are so ham fisted that they cannot even point out a rules violation with suspending an actress's account.
> Or of course that's just all #fakenews - the abusive behaviour?
The whole "threatening nuclear war violates" as abusive behavior is a losing argument. The President doesn't like a lot of people and doesn't conceal that information. The idea any of his tweets go beyond hurting someone's feelings is pretty absurd. In contrast, Twitter allows actual threats of assassination without consequence. Nevermind all the threats of rape, etc. directed at women on Twitter without any answer. Look at any popular, conservative women's timeline and see all the crap the abuse council and twitter support lets through. Actual threats versus hurt feelings shows what should have action taken.
All of this is just some people's desire to remove a direct outlet from groups of people. It has nothing to do with actual threats.
This sounds somewhat patronizing (the first part in particular), but falling victim to confirmation bias is something we all do. When ideas fit our own personal biases, we tend to become much less critical of them. For instance animal testing, when visible, is something very few people can emotionally accept. And there have been countless hoaxes [1] where people will share an image of an animal in one context (such as a rabbit suffering from severe hair less / skin damage at a veterinarian) and then claim it's an image of the result of a named shampoo company testing their products on animals. It gets people riled up and interested in stopping animal testing, but the problem is that it's completely fake. This is an obvious example but the exact same is true of words themselves and it spreads into everything -- most notably politics.
It's in many ways bizarre that we don't deal with this issue as a part of basic education. Imagery or messages designed to spark an emotional response are very effective against people who are not aware of what's happening. At the same time, they can be rendered far less potent by simply educating people about these tools of manipulation and giving them a wide swath of examples. In today's ever more connected world, with ever more people looking to shall we say 'utilize' other people, the complete neglect of this social skill in education today is perplexing.
[1] - https://speakingofresearch.com/2017/05/16/context-matters-ho...
identify verification when you create your account. charge a small fee when you create your account.
>Consequently, I think instead of trying to solve the issue of bots, it's more important to start educating people that just because they read something on the internet does not mean it's true
now THERE is a problem that cannot be solved!
It's impossible to take everything critically, and honestly few will. So the entity with the largest bot army still has the longest propaganda lever.
Twitter and the other social networks know who is a bot, or else they haven't bothered looking. Something needs to force them to act.
I do see one mechanism: the bots go too far, and users don't want to be on a platform where they just interact with bots, so they go to more curated places to get their fill. So the business health of the platform depends on having trust. FB has a leg up on this since your friend list probably has people you've actually met. There the problem is your gullible friend forwarding you crap. A deputized bot, if you will. No level of education helps there.
When I speak of bots, I am implicitly speaking of the inevitable adaptions to any sort of attempt to crack down on them. I do agree with you that right now many bots can be detected pretty easily. But that's largely because they have no reason to disguise the fact that they're a bot. In many ways, I think the current system is more desirable. As bots progress to actually trying to emulate human behavior it's going to result in the sort of paranoia you see on many forums today where individuals call one another 'shills' as a means of expressing disagreement. And ultimately, I do not think it will be at all difficult to pass a heavily crippled turing test of 140 character unidirectional messaging.
But of course Twitter has no obligation to ban them.
If anyone is interested, here is an example of probably bot account [1]. It was registered in Nov 2013 and has posted 27 000 tweets and retweets which is 18.5 tweets per day on average. Another account [2] has a random user id and has posted 2000 tweets in last 3 months.
And by the way I don't think that bots influenced election results. I watched the debates and Hillary's position was very weak.
There's a lot of misunderstanding about what these bots are capable of... and I keep seeing top ranking comments on Reddit talking about these bots daily.
The media seems to be spreading the idea they are autonomous agents automatically posting content on politically useful threads, with bot voting rings pumping them up. When in reality they still rely heavily on a human-intensive process, requiring lots of manual targeting and copywriting, and Reddit/Twitter/etc are very good at detecting voting-rings, having been perfecting algorithms to detect them for over a decade.
And do people think Russia or fringe right-wing groups have some super-effective autonomous bots that weren't also available to the best paid US consultants?
If these fake Twitter accounts, usually bots only followed by other bots, were understood in terms of the hundreds of millions of real users on Twitter, the billions spent by both real parties, and within the context of the technical limitations of bots, it wouldn't seem so scary.
> A researcher finds 1,600 bots tweeting extremist posts in U.S. elections also spread anti-Macron sentiment in France
A second quote from the article shows a much larger bot population.
> Research from the University of Southern California and Indiana University shows that 9 to 15 percent of active Twitter accounts are bots.
A bot posting articles from your favourite news site is actually useful. A bot which prints out Trump's tweets, burns them and then tweets out the video [@burnedyourtweet] is mildly amusing (and some pretty smart engineering). A bot which automatically replies to every tweet by $personality with a meme is usually an 'orrible troll, but occasionally amusing and often owned by someone who's very convinced the First Amendment applies to Twitter. A retweet bot which does a plausible impression of an angry Trumper might be indistinguishable from the real thing without investigating its network, and I don't think eliminating the real thing would be a great move for Twitter to make.
The other reason is that Twitter's review process is more or less completely random in its responses (I'm reminded of the girl that reported literally the same abusive tweet directed at her from three troll accounts, and received a different response for each one)
This community driven ranking seems like the best solution to low quality bots posting. A modding driven centralized editing process comes with plenty of issues, as we've seen on Reddit. And yes Bot's provide plenty of value, far outweighing some niche political accounts making the rounds.
I have plenty of fake 'bot' followers, and 99.9% of their followers are other bot. It's largely a bunch of bots tweeting to other bots.
The main risk is hashtags being gamed and people buying voting rings/followers... which has been a problem on the internet forever.
There were about 400,000 bots posting political messages during the 2016 U.S.
presidential election on Twitter, according to a research paper by Emilio
Ferrara, an assistant professor at the University of Southern California.> He told Bloomberg that he has discovered that the same group of 1,600 bots tweeting extremist right-wing posts in the U.S. elections also posted anti-Macron sentiment during the French elections and extremist right-wing content during the German elections this year.
So of those 400k political bots, 1600 were identified as talking about fringe topics.
Not to mention it's not clear how this fits into the bigger picture? What percentage of ALL of Twitter accounts are bots? How many of those bots are active? How many actually have an audience of real people, not just other bots? For ex: 400k fake accounts, with few real followers, out of hundreds of millions of real users talking politics may not seem as significant...
Not saying it's not a 'problem', but it needs to be understood in context, especially considering the constant news coverage and outrage.
Read it again. Of those 400k bots, 1600 were identified talking about the US elections, anti-Macron sentiment, and extremist far-right topics. All three topics. That could be one person/group reusing bots.
Additionally, hundreds of millions of real users talking politics is unlikely. Twitter has 300 milliom active users per month, 9-15% of them bots, the US election had 120 million voters. Hundreds of millions would be nearly everyone.
Fair enough. So if 10% of voters used Twitter, that means 3% (400k) were bots, and of those a tiny percent were fringe/far-right.
Nothing posted is proof of that. The 1600 number is interesting as it means one entity was trying to influence elections in three countries. It is not the number that were far-right, that number was not in this article.
As context is supposed to be a needed thing, you could have just found the research paper rather than perform guestimates math. Fifteen percent of users tweeting about the election were bots, twenty percent of tweets, and about three Trump bots for every Hilary bot.
http://firstmonday.org/ojs/index.php/fm/article/view/7090/56...
Those numbers aren't ads/posts, they're dollars. In September, Facebook confirmed that Russian operatives spent $100,000 on psyops advertising.
Keep in mind that (1) $100,000 is 15-25 million impressions reaching millions of highly-targeted people (Cambridge Analytica is now under investigation by the HPSCI), all amplified by the targets' engagement, and that (2) FB is sharing only 100% confirmed activity discovered during very early, self-investigated findings.
In that context, how could $100,000 of FB ads (0.001%) possibly be relevant?
Speculating that it might possibly be relevant if we hypothetically were to discover it was 100x bigger than we currently know, to me that defeats the whole argument.
"Big if true!" is the rallying cry of the whole Russiagate fiasco. "Big if Bigger!" is just the next contortion.
[1] - https://www.cbsnews.com/news/election-2016s-price-tag-6-8-bi...
That said, I think it's possible that we actually are doing the same thing.