bots – Techdirt (original) (raw)

Crawlers And Agents And Bots, Oh My: Time To Clarify Robots.txt

from the perplexing dept

Perplexity is an up-and-coming AI company that has broad ambition to compete with Google in the search market by providing answers to user queries with AI as its core technology.

They’ve been in the news because their news feature repurposed content published on the Forbes website in an investigative article, which severely annoyed the Forbes editorial staff and media community (never a good idea) and led to accusations from Forbes’ legal team of willful copyright infringement. Now Wired is reporting that Perplexity’s web hosting provider (AWS) is investigating their practices, focused on whether they respect robots.txt, the standard governing the behavior of web crawlers (Or is it all robots? More on that later.)

We don’t know everything about how Perplexity actually works under the hood, and I have no relationship to the company or special knowledge. The facts are still somewhat murky, and as with any dispute over the ethics or legality of digital copying, the technical details will matter. I worked on copyright policy for years at Google, and have seen this pattern play out enough times to not pass judgment too quickly.

Based on what we know today from press reports, it seems plausible to me that the fundamental issue at root here, i.e. what is driving Perplexity to dig its heels in, and where much of the reporting seems to cite as Perplexity’s fundamental ethical failing, is what counts as a “crawler” for the purposes of robots.txt.

This is an ambiguity that will likely need to be addressed in years to come regardless of Perplexity’s practices, so it seems worth unpacking a little bit. (In fact similar questions are floating around Quora’s chatbot Poe.)

Why do I think this is the core issue? This snippet from today’s Wired article was instructive (Platnick is a Perplexity spokesperson):

“When a user prompts with a specific URL, that doesn’t trigger crawling behavior,” Platnick says. “The agent acts on the user’s behalf to retrieve the URL. It works the same way as if the user went to a page themselves, copied the text of the article, and then pasted it into the system.”

This description of Perplexity’s functionality confirms WIRED’s findings that its chatbot is ignoring robots.txt in certain instances.

The phrase “ignoring robots.txt in certain instances” sounds bad. There is the ethical conversation of what Perplexity is doing with news content of course, which is likely to be an ongoing and vigorous debate. The claim is that Perplexity is ignoring the wishes of news publishers, as expressed in robots.txt.

But we tend to codify norms and ethics into rules, and a reasonable question is: What does the robots.txt standard have to say? When is a technical system expected to comply with it, or ignore it? Could this be rooted in different interpretations of the standard?

First a very quick history of robots.txt: In the late 80s and early 90s, it was a lot more expensive to run a web server. They also tended to be very prone to breaking under high loads. As companies began to crawl the web to build things like search engines (which requires accessing a lot of the website), stuff started to break, and the blessed nerds who kept the web working came up with an informal standard in the mid 90s that allowed webmasters to put up road signs to direct crawlers away from certain areas. Most crawlers respected this relatively informal arrangement, and still do.

Thus, “crawlers” has for decades been understood to refer to systems that access URLs in bulk, systems that pick which URLs to access next based on a predetermined method written in code (presumably why it’s described as “crawling”). And the motivating issue was mainly a coordination problem: how to enable useful services like search engines, that are good for everyone including web publishers, without breaking things.

It took nearly two decades but robots.txt was eventually codified and adopted as the Robots Exclusion Protocol, or RFC 9309, by the Internet Engineering Task Force (IETF), part of the aforementioned blessed nerd community who maintain the technical standards of the internet.

RFC 9309 does not define “crawler” or “robot” in the way a lawyer might expect a contract or statute to define a term. It says simply that “crawlers are automatic clients” with the rest left up to context clues. Most of those context clues refer to issues posed by bulk access of URIs:

It may be inconvenient for service owners if crawlers visit the entirety of their URI space. This document specifies the rules […] that crawlers are requested to honor when accessing URIs.

Every year the web’s social footprint expands and we increase the pressures put on robots.txt. It’s begun to solve a broader set of challenges, beyond protecting webmasters from the technical inconveniences of bulk access. It now increasingly arbitrates massive economic interests, and now the social and ethical questions AI has inspired in recent years. Google, whose staff are the listed authors of RFC 9309, has already started thinking about what’s next.

And the technology landscape is shifting. Automated systems are accessing web content with a broader set of underlying intentions. We’re seeing the emergence of AI agents that actually do things on behalf of users and at their direction, intermediated by AI companies using large language models. As OpenAI says, AI agents may “substantially expand the helpful uses of AI systems, and introduce a range of new technical and social challenges.”

Automatic clients will continue to access web content. The user-agent might even reasonably have “Bot” in the name. But is it a crawler? It won’t be for the same purpose as a search engine crawler, and not at the same scale and depth required for search. The ethical, economic, technical, and legal landscape for automatic AI agents will look completely different than for crawlers.

It may very well be sensible to expand RFC 9309 to apply to things like AI agents directed by users, or any method of automated access of web content where the user-agent isn’t directly a user’s browser. And then we would think about the cascading implications of the robots.txt standard and its requirements. Or maybe we need a new set of norms and rules to govern that activity separate from RFC 9309.

Either way, disputes like this are an opportunity to consider improving and updating the rules and standards that guide actors on the web. To the extent this disagreement really is about the interpretation of “crawler” in RFC 9309, i.e. what counts as a robot or crawler and therefore what must respect listed disallows in the robots.txt file, that seems like a reasonable place to start thinking about solutions.

Alex Kozak is a tech policy consultant with Proteus Strategies, formerly gov’t affairs and regulatory strategy at Google X, global copyright policy lead at Google, and open licensing advocate at Creative Commons.

Filed Under: agents, ai, bots, crawling, generative ai, robots.txt
Companies: perplexity

ExTwitter Users Getting Fed Up With The Crypto Spam And AI Bots Elon Promised To Clean Up

from the turns-out-content-moderation-is-difficult dept

Among the various promises that Elon made regarding his takeover of ExTwitter, was that he was there to clean up the spam and bot problem. He seemed to think that the previous regime had fallen down on the job, and that somehow he would have the magical answer to dealing with such things.

About that.

Originally, Elon seemed to think that changing Twitter’s verification system into a subscription service would get rid of the bots. That did not work. More recently, he’s shifted into making anyone who wants to post anything to Twitter to have to pay a nominal amount as his solution.

All of this assumes, incorrectly, that it’s not worth it for scammers and spammers to pay tiny bits to flood Elon’s playground with shit.

And flood it, they are.

A report from Bleeping Computer notes that ExTwitter has become completely overwhelmed with crypto scam ads, and most of them are coming from accounts paying Elon his cut. And even Elon’s biggest supporters are getting sick of it.

And it seems clear that it’s worth it to scammers to pay 8/monthforaccesstotheabsolutegulliblefucksonExTwitter.AsBleepingComputerhighlightedlastmonth,oneofthesecryptodrainerscamsthathasbeenregularlyadvertisingonExTwitter[wasabletosteal8/month for access to the absolute gullible fucks on ExTwitter. As Bleeping Computer highlighted last month, one of these crypto drainer scams that has been regularly advertising on ExTwitter [was able to steal 8/monthforaccesstotheabsolutegulliblefucksonExTwitter.AsBleepingComputerhighlightedlastmonth,oneofthesecryptodrainerscamsthathasbeenregularlyadvertisingonExTwitterwasabletosteal59 million from suckers on ExTwitter via purchased ads:

On X, better known as Twitter, advertisements for MS Drainer are so abundant that ScamSniffer reports they account for six out of nine phishing ads on their feed.

Notably, many of the scam ads on X are posted from legitimate “verified” accounts that carried the blue tick badge when the ad was shown.

The account MalwareHunterTeam is out there finding more and more such scam ads on ExTwitter. Here are just a few:

And more and more and more.

Meanwhile, Elon’s other big “innovation” to try to stop the bots was to change the way the API worked so that it charged ridiculous fees to use. Of course, all that’s done is driven away the useful bots, but left the scam bots free to be.

Which brings us to the other story demonstrating Elon’s absolute failure to deal with bots on the platform. Boingboing details how Parker Malloy has found that there appear to be a shitload of fake bot accounts that are clearly running off of ChatGPT. And you can tell that by simply searching for the phrase “goes against OpenAI’s use case policy.” You find tons and tons of tweets using that phrasing, as it is clearly coming from a bot powered by ChatGPT, but where whoever set it up didn’t think that OpenAI would reject their query.

Post by @parkermolloy View on Threads

And, look. Fighting spam and bots is a big challenge. And if Elon had approached this with even the slightest humility, you might feel bad for him. But instead, he insisted, without knowledge, that Twitter’s previous management was failing to take the problem seriously, was lying about how much spam was on the platform (even though that was only because he couldn’t understand how Twitter was reporting things), and that somehow he would have the singular solution to solve it.

Instead, he fired basically anyone who knew anything about fighting spam, put in place braindead stupid solutions that anyone with any experience in the field would tell you wouldn’t work… and then made the problem way, way, way worse.

No wonder Elon is now moving on to trying to blame “DEI” for anything bad that happens in the world (someone should tell him that his own companies, Tesla and SpaceX, both advertise their DEI efforts, but alas…)

Filed Under: ai, bots, chatgpt, crypto spam, elon musk, scams, spam, verification
Companies: twitter, x

The Great Paywall Of Musk Will Consist Of $1/Year To Tweet

from the no-thank-you dept

You will recall that, last month, Elon Musk mentioned in passing that he’d decided the only way to stop bots and spam on Twitter (which he’d already claimed to have stopped a few times earlier) was to paywall the entire site with “a small monthly payment for use of the X system.” This statement apparently caught CEO-in-name-only Linda Yaccarino entirely by surprise, as she seemed unaware of the issue when asked about it live on stage a couple weeks later.

Last night, Fortune’s Kylie Robison has broke the news that the plan is to charge new users $1/year in order to access the basic features of exTwitter:

X, the platform formerly known as Twitter, will begin charging new users $1 a year to access key features including the ability to tweet, reply, quote, repost, like, bookmark, and create list, according to a source familiar with the matter

X owner Elon Musk has long floated the idea of charging users $1 for the platform. During a recent livestreamed conversation with Israeli prime minister Benjamin Netanyahu last month, Musk said “It’s the only way I can think of to combat vast armies of bots.”

Soon after her article came out, exTwitter “Support” tweeted a confirmation, officially naming this new paywall “Not A Bot” and saying that it’s already being tested in New Zealand and the Philippines.

The webpage detailing the new program says that when new users sign up in those two countries, they’ll be given options for what “Subscription Plan” they want, and if you don’t pay up, you’ll be read only.

So, as fees go, 1/yearisobviouslynotmuchatall.But,um,ifyou’reaspammer,andyoustillwanttospam,would1/year is obviously not much at all. But, um, if you’re a spammer, and you still want to spam, would 1/yearisobviouslynotmuchatall.But,um,ifyou’reaspammer,andyoustillwanttospam,would1/year be much of a deterrent? But it does seem likely to deter more casual users who just want to try exTwitter. Which means that it will most likely slow down user growth without doing much to stop spam.

And that’s the fundamental problem here. There’s no way to charge users in a manner that it’s more worthwhile for users to pay than for spammers. The most problematic spammers are doing it for the money. And as long as they can make more than $1/year it’s still going to be worth it to them. Perhaps it’s stops a total flood of purely nonsense spam trolls, but… it seems highly unlikely to stop the actual worst of the spammers.

But, it does seem like the kind of friction that is pretty likely to kill off more casual usage, which the site relies on to grow.

And while perhaps it’s not a huge deal if it really is only going to be for “new” users (so long as you believe the site is fine with the users it currently has), how long does anyone really believe that’s going to last? If Musk remains hard up for cash at exTwitter, you have to expect that the fee will be expanded to everyone over time. As for testing it in just those two countries, I’m sure people in those countries can figure out how to proxy or VPN their way around the geogating. Which means, whatever “results” they have of this test are not going to be all that interesting.

These are, of course, all ideas that many people in the space could tell Elon, if he hadn’t fired/banished/ignored all the experts over the past year, leading to him reinventing every mistake from “first principles.”

Filed Under: bots, charging, elon musk, new zealand, paywall, philippines, social media, spam
Companies: twitter, x

Elon, Who Insisted He Had Solved The Spam/Bot Problem, Now Says He Will Paywall Twitter To Stop Spam/Bots

from the admission-fee-to-the-public-square dept

So, remember when Elon first announced his plans to buy Twitter, and he claimed that it was “the de facto town square” and his top priority was to eliminate spam and scam bots?

“A top priority I would have is eliminating the spam and scam bots and the bot armies that are on Twitter. They make the product much worse.”

He tweeted out that he was going to “defeat the spam bots or die trying!”

Of course, a month later, when he decided that he wanted out of the binding deal he signed that had no outs, he suddenly started pretending that he could get out of the deal because there were more spam bots than he originally believed. That made no sense, was wrong about the data, and also was not even remotely an effective legal strategy, as he ended up being forced to pay the price he had promised.

Of course, since then, Elon has repeatedly claimed that he’s magically fixed the problem of spam bots on the platform. Back in December of last year, Musk claimed that he had a plan to shut down bot IP addresses and that would magically solve the bot problem that he insists the company failed to fix before the genius that is Elon Musk showed up.

Musk took a bit of a victory lap about how he had defeated the bots, only to find out that… his actions did not do a damn thing to stop spam bots, which are now more prevalent than ever on the site. Of course, in the lore of Musk, everyone must pretend that the earlier announcements he’s made that didn’t pan out didn’t exist, so no one ever seems to follow up on any of these claims, or the fact that Musk has repeatedly promised that things like his $8/month verification problem will magically solve the bot problem, when it only served to further enable scammers to prey on people by pretending to be legit.

These are all entirely expected results that Elon could have understood if he hadn’t fired everyone who actually understood the process of fighting spam. None of this is secret. However, with Musk, he always has to ignore expertise and go with his gut, even when his gut is ill-informed.

The latest potential move he’s talking about is making everyone have to pay a small monthly fee to use exTwitter.

Among other things, Musk said the social network is “moving to having a small monthly payment for use of the X system” in order to combat “vast armies of bots.”

As always with Elon, it’s difficult to know what he will actually do and what is him just talking nonsense. Having followed the guy and his announcements for quite some time now, it appears that he has a rather Trump-like ability to say ridiculously stupid stuff that he 100% believes in the moment, and the question of whether or not it actually goes anywhere comes down to two factors: (1) if someone close to him who knows how absolutely stupid the idea is can talk him out of it or (2) if someone close to him who knows how absolutely stupid the idea is can distract him with some new shiny object to take his attention off this idea for long enough that he forgets about it.

If either of those things happen, the plan to paywall “the de facto public square” to defeat the bots Musk already insisted he had defeated may just slowly fade away and be forgotten. If he does implement it, however, it will almost certainly be a disaster for exTwitter, and will lead to a much bigger rush to other sites.

The thing is, like so many of Musk’s ideas for exTwitter, this one is so unquestionably, obviously stupid and counterproductive that it just lays bare how little Musk understands about social media and the internet. I have my doubts as to whether or not he will actually go through with it, but I also have a morbid fascination with watching what would happen if he does…

Filed Under: bots, elon musk, monthly fee, paywall, spam
Companies: twitter, x

Elon’s Twitter Kills Off Many More Useful Bots

from the good-bot-bad-bot-elon-bot dept

One of Elon’s big promises when he took over Twitter was that he would get rid of spam bots. So far that’s been a huge fucking failure. That’s from the Wall Street Journal, which has been generally supportive of Musk’s tenure at Twitter. But, the article makes it clear that Musk has totally failed to make any dent in the fight against spam bots. The article quotes a bunch of experts and researchers pointing to various studies and reports all saying that the amount of spam on Twitter doesn’t seem to have changed much. And the article also notes that Musk’s claims that his Twitter Blue fake verification plan (which Musk insisted was key to stopping spam bots) has actually made the problem worse:

Some evidence indicates that Twitter Blue has actually added to the problem of fakes and bots because the way it changed the verification process has left many users unsure which accounts are real and which are fake.

Subscribers can now purchase check marks that Twitter used to reserve for accounts the company had deemed authentic and notable. Some famous users who previously had blue check marks now don’t, leaving those accounts vulnerable to imitators.

“Users are worse off in trying to delineate trustworthy and not trustworthy accounts,” said Princeton’s Mayer, whose study indicated that most U.S. adults don’t understand the platform’s criteria for assigning blue check marks to profiles.

And it’s been clear throughout this process that Musk has never even attempted to understand how any of this works, and makes his decisions almost entirely based on his own, somewhat distorted, view of how the platform works, which is extremely different as one of its most popular users.

And, of course, the prevalence of spammers on Twitter is so obvious now that when one of Musk’s most loyal fans tweeted that spam seemed to have disappeared on the site, multiple users showed up in the replies typing something along the lines of “my Facebook has been hacked, bitcoin, doge, nft” and the spammers went so crazy that the Musk fan deleted his tweet. As for the claim that Twitter Blue “verification” would somehow solve the problem, uh, no. Spammers seem happy to pay $8/month to spam people. Over on Bluesky, the user “Kilgore Trout” recently posted pages and pages of “Twitter Blue” accounts that the account had blocked, all of which had the blue check mark:

Of course, one of the parts that Musk really doesn’t seem to understand is the difference between bot spammers and useful bots. Because of his complete lack of comprehension on this topic, he seemed to think that beyond using Twitter Blue to stop bots (failure), his other big move would be to start charging for the API.

The assumption there seemed to be that all “bots” that used the API were “spam.” But, that’s never been true. Indeed, many of the bots that use the API were creating useful tools that made Twitter better. When he first announced his plans to start charging for the API and getting rid of the “free” tier, he seemed taken aback and surprised when people pointed out that there were tons of useful bots on the site, leading him to hastily announce that they’d create a free API for “good bot content.”

But, of course, there were no details, and the arbiter of “good bot content” wasn’t based on any principles, but pretty much seemed to be what Elon decided was good. Over the last few months, there have been a few different “bot apocalypses” in which suddenly a bunch of bots that used the Twitter API went offline (this includes us, by the way, as our auto-posting of Techdirt stories to Twitter is no longer allowed, since apparently we’re not “good bot content.”)

But, over the last few days, there’s been yet another bot apocalypse, as a bunch of “good bots” posting fun content started disappearing from the site.

Over the past 24 hours alone, Twitter suspended API access for numerous bot accounts that post photos of animals. Far from being a nuisance, accounts of this type tend to rack up large fanbases. For instance, @PossumEveryHour, which posts photos of possums for its more than 500,000 followers, announced it would be shutting down after losing API access on Friday night. @hourlywolvesbot, which tweets wolf pics for its more than 173,000 followers, also announced it would no longer be posting to Twitter for the same reason.

Other animal picture-posting bot accounts that have announced they’d no longer be able to post on Twitter include @CorgiEveryHour, @HourlyCheetahs, and @HourlyLynxes.

“This app has violated Twitter’s Rules and policies,” reads the message provided to the suspended bot accounts in Twitter’s developer portal. “As a result, it can no longer be accessed.”

As the article notes, some of the bots have been able to come back, as the issue for them seemed to be that they were still using the old API, and Twitter finally shut that down. So, for those accounts, they’re able to get started again by switching to the new API. But that’s not true for all of them, as many are just totally being shut out, unless people complain loudly enough and Elon decides that the bot is good enough to return.

Some users have found a solution for the bot accounts. It appears that for some bots, the issue is just that Twitter is now getting around to shutting down their old free API tiers. This means that some bots could continue to run, albeit with less regularity, via Twitter’s new extremely limited free tier that was set up for those “good” bots. They would need to manually setup the account once again on the new Twitter API.

But, that’s not the case for every bot account, as some have found that they still need to shut down unless they pay for API access. And since most of these accounts don’t make their creators any money, few are willing to do so.

And it hasn’t only been the animal-posting bot accounts that have been affected. For example, the popular @MakeItAQuote account, which has more than 623,000 followers, was one of the first taken down in this purge when its API access was suspended last week. The account would automatically create a quote image of a user’s tweet when someone mentioned it in the reply to the post.

Hilariously, as the Mashable article notes, it’s possible that all of this is in response to Elon discovering that the bot account “ExplainThisBob,” was really pushing some sort of crypto shitcoin. The account, which had built up quite a following by using some sort of AI system to “explain” tweets with summaries that were often a mess, had been praised just weeks ago by Elon himself, who claimed “I love Bob” after Bob had summarized a discussion about Twitter cutting off API access:

But, then once he realized that the Bob account was pushing a crypto shitcoin, Elon said it had to be suspended, leading the bot to beg for its life, unsuccessfully.

Again, the only consistency here are Elon’s whims. And, let’s be quite clear: it’s his site. He can do whatever the fuck he wants with it. But, what’s hilarious is that he kept insisting that one of the key reasons he took over the site was to stop the inconsistent and arbitrary moderation policies (back when there were actual policies) and he’d bring about more consistent policies, mostly around letting everything go.

Except, apparently, if you post bots of cute animals.

Filed Under: api, bots, content moderation, elon musk, explainthisbob, possumeveryhour, scams, spam, useful bots
Companies: twitter

Elon Promises A Free API For ‘Good Bot Content,’ Again, Demonstrating He Has No Idea How Any Of This Works

from the whims-are-not-policy dept

It’s been clear since the takeover, that Elon’s running Twitter entirely based on his fleeting and oft-changing whims. The weird decision last week to suddenly, with one week’s notice, remove the free tier for Twitter’s basic API, has create a bit of an uproar, as tons of tools, services, and useful bots made use of it. Many have been posting farewell messages on Twitter, leading Musk (as he seems to do all too often) to announce a policy change in a reply tweet. He did this when he rolled back his bizarrely stupid policy that you were no longer allowed to link to other social media (a policy so obviously stupid, that only Musk’s mother would defend it). Musk rolled that one back in a reply tweet — meaning a tweet that very few people would see, because they don’t show up nearly as much.

Here, Musk responded to a tweet from the automated @PepitoTheCat account, which pointed out his account would have to be shut down under this new policy. The new policy, according to Musk in this reply tweet (which got many fewer views that his regular tweets) is that:

I guess we could give all Verified users access to the API for posts like this

Responding to feedback, Twitter will enable a light, write-only API for bots providing good content that is free

Really.

So… the policy might be that “all verified users” get access to the API… but only for “posts like this.” Or… it could be that a “write-only API” (don’t even get me started on how nonsensical that is) for “bots providing good content that is free.”

The point is: these are not policies. These are brain fart whims. This is no way to run a business with many millions of users.

Of course, it’s reminiscent of the naivety that Musk has demonstrated about all of this before. Like when he declared that content moderation was simple: you just delete the “wrong and bad” stuff, and support the “good” content.

Okay, genius: define “good.” Define “good” in a manner that your remaining team of flying monkeys can put it into practice without having to constantly consult your tweet replies to see if they’re doing it right.

The process of content moderation on a competent trust & safety team involves crafting a policy that can be understood by the team in charge of carrying it out. Creating a truly exceptional policy would be one that most users can understand.

None of that is happening here. Musk’s random “well, if it’s good” they can get some sort of weird sorta API access is not fixing anything. This is what good trust & safety teams actually do, and what they’re skilled at doing. It’s a constant challenge, of course, because as you write policies, you’ll constantly be running up against exceptions and things that challenge the policies. But saying “make an exception for ‘good’ bots” is worse than useless.

It’s simply reinforcing just how risky and ridiculous it is for anyone to build anything that relies on Twitter today. Who knows what nonsense will come from a random reply Tweet to a cat meme tomorrow?

Meanwhile, I’ll just note that over the weekend, I discovered three new services being built on the Mastodon API that are replacing things I used to use Twitter for. One of which, FeedSeer, is basically just like TweetShelf, an incredible Twitter tool that seems likely to need to shut down thanks to these changes.

Filed Under: api, bots, content moderation, elon musk, good content
Companies: twitter

Multiple Former Twitter Employees Note That Musk’s New Favorite Tool, Polls, Are Easily Gamed By Bots

from the but-of-course dept

Rolling Stone has a fun article quoting multiple former Twitter employees highlighting that polls are the least secure tool on the platform, and are regularly open to manipulation by bots:

“Polls are more prone to manipulation than almost anything else [on Twitter]. It’s interesting, given his [Elon’s] use of polls,” he added. Several other ex-Twitter employees gave similar assessments.

This seems particularly notable for two reasons: (1) Musk’s sudden reliance on polls for making big content moderation decisions, and (2) his formerly professed (though of questionable seriousness) claims about concerns regarding bots on the platform.

On point one, we already discussed the ridiculousness, and lack of seriousness, in using easily gamed polls as a tool for content moderation. While supporters like to argue it’s “democratic,” it has none of the actual hallmarks of integrity around the “voting.” And this report regarding the manipulation just makes that even clearer:

“When someone says. ‘Oh we must be protecting polls, right?’ No, we’re not,” the former Twitter employee told Rolling Stone.

In the years since the feature debuted, a small industry of spammers has cropped up to offer services manipulating the results of a Twitter poll with inauthentic accounts. The spammers allow users to buy votes in chunks, some offering 15,000 votes on a given poll for a little over $130 or smaller responses for just 19 cents a vote with “guaranteed fast delivery” that’s “100% Confidential.”

For what it’s worth, the Rolling Stone article perhaps gives a little too much credence to the idea that Musk ever seriously considered “bots” a problem on Twitter. It was always clear that it was a pretense to try to get out of the deal. So the fact that he pretended to care about bots on the platform for a few months shouldn’t be taken to mean he really believes it’s a problem. Especially right now when he desperately wants to show growth to woo back advertisers who have abandoned ship.

The Rolling Stone piece does a nice job also highlighting how Musk’s recent claims of big increases in the mDAU may also be facing the same issue as the polls: a lack of staff manually removing spammers:

But it’s not clear how much of that claimed growth is authentic. Asked whether those numbers could be inflated by spam accounts, the former Twitter staffer told Rolling Stone: “No fucking doubt.”

“Think about it: On any given week, [the security] team removed millions of accounts manually,” the source said.

Of course, on Wednesday, Musk publicly claimed that the site was removing a bunch of spammers:

Of course, somewhat hilariously, the purge ended up killing a bunch of high profile legitimate accounts. Early on, there were reports of some high profile “left leaning” critics of Musk who were removed leading to claims that the Muskian Twitter was dealing in “anti-left bias,” but as with the years of false claims under the old regime of “anti-conservative bias” the reality appeared to be much more mundane: the impossibility of doing content moderation well at scale. Indeed, some of the other accounts that were suspended included Elon Musk’s most vociferous number one fan.

Turns out content moderation, including dealing with spam and bots, is, you know, not easy.

Filed Under: bots, elon musk, polls, spam
Companies: twitter

Creator Of Botometer Goes On Media Tour To Explain Why Elon Musk’s Claims About Bots (Using Botometer) Are Meaningless

from the convince-me-musk-is-not-a-bot dept

As you may recall, in his response to Twitter’s lawsuit trying to force him to fulfill the terms of the purchase agreement he made, Elon Musk relied on the findings of a tool called Botometer to argue that there were more bots on Twitter than Twitter was claiming. Again, I have to remind everyone, as much as Musk keeps insisting this case is about bots and spam, the actual case has nothing at all to do with bots or spam, and if you think it does, you’ve been lied to.

However, Musk is doing everything he can to confuse people, including the judge, into believing the case is about bots and spam. And, in doing so, he feels that he needs to convince people that there are a lot of bots and spam on the platform (he claims more than Twitter admits to, but he — and many others — are totally misrepresenting what number Twitter is actually reporting). In his attempt to confuse everyone, in the lawsuit, Musk cited results from Botometer — which made tons of experts in the space laugh. Botometer is a fun toy to play around with, but no one takes it seriously as an actual tool to determine how many bots are on Twitter.

Including Botometer’s creator.

In the wake of Musk putting all of his eggs in Botometer’s basket, it appears that a creator of Botometer has been making the media rounds pointing out that Musk is a fool.

A couple weeks ago, grad student Kai-Cheng Yang from Indiana University was interviewed by Yahoo Finance, and helped debunk some of Musk’s talking points. First, he notes that what Twitter is reporting to the SEC as bots in its monetized daily active users is a different thing than Botometer tries to analyze. There is some overlap, but they’re looking at different things:

I think Twitter has made it clear that they are focusing on spam and the false accounts. And my understanding of their definition about spam accounts is that those kind of accounts would send repeatedly all different kinds of information, trying to promote some website or some product or some cryptocurrency to people, kind of annoying, right? But you can achieve those kind of goals through bots, of course. But also you can have real people control those accounts. In my opinion spam accounts has an overlap with social bots, which is what we detect. But also it’s not entirely the same thing.

He also pours some cold water on an oft-cited claim from five years ago, made by the team behind Botometer, that it believed between 9% and 15% of Twitter was bots. As he notes, that was a long time ago — Botometer has changed a ton since then and measures things differently, and (perhaps more importantly) Twitter’s own approaches to dealing with bots and spam has changed drastically as well.

Yes. So our group did those kind of estimations back then. But I do want to mention that we’ve been upgrading our tools constantly. So the Botometer today is different from what we have before, right?

And also the situation on Twitter has been changing quickly, because Twitter also, they have been doing a lot trying to remove bots and other inauthentic accounts. So actually, I think they drive to the bad actors to change their behavior, to change their accounts. So I am not sure that estimation is still accurate today, unfortunately.

Finally, when asked about Musk’s use of their tool to calculate bots, Yang basically says that users of his tool can effectively tweak the threshold to churn out any number they want for bots, so without knowing that (which Musk hasn’t disclosed), you can’t tell whether or not his estimate is reasonable:

Our tool works, if you give it account, it will give you a score. If the score is higher, it means the account is bot-like. If the score is low, it means it’s human-like. But it’s a score. So in order to have a number of percentage of bots on Twitter, you have to choose a threshold, right?

And that’s, I don’t know how Elon Musk did it. And technically you can choose any threshold you want and to get any result you want. So that’s my understanding right now. Elon Musk didn’t make it clear how they choose this threshold to me.

This is all sorts of funny, because throughout this whole thing Musk keeps complaining that Twitter hasn’t given him the relevant thresholds on how it scores spam (except, as noted at the bottom of this article, it has), and now the creator of the tool that Musk is relying on says that Musk is actually hiding the most important part of his own analysis.

The very next day, Yang also did an interview with CNN where he was even more aggressive in highlighting how absurd it is that Musk is relying on his tool.

“To be honest, you know, Elon Musk is really rich, right? I had assumed he would spend money on hiring people to build some sophisticated tool or methods by himself,” Yang told CNN Business Monday. Instead, Musk opted to use the Indiana University team’s free, publicly available tool.

He also reminded CNN that spam and bots are not the same thing, and all his tool does is try to determine automated accounts — many of which are legit and not spam.

He further delved into the important threshold setting question that basically allows anyone using his tool, Musk or not, to falsely imply something by fiddling with the threshold:

“It’s tempting to set some arbitrary threshold score and consider everything above that number a bot and everything below a human, but we do not recommend this approach,”

And then, last week, he did yet another interview, this time with the BBC, this time being even more aggressive in saying that Musk’s use of his tool is nonsense.

Using the tool, Mr Musk’s team estimated that 33% of “visible accounts” on the social media platform were “false or spam accounts”.

However, Botometer creator and maintainer, Kaicheng Yang, said the figure “doesn’t mean anything”.

As Yang explained, again, Musk doesn’t say what threshold he used, and that allows him to say whatever he wants.

“In order to estimate the prevalence [of bots] you need to choose a threshold to cut the score,” says Mr Yang.

“If you change the threshold from a three to a two then you will get more bots and less human. So how to choose this threshold is key to the answer of how many bots there are on the platform.”

Mr Yang says Mr Musk’s countersuit does not explain what threshold it used to reach its 33% number.

“It [the countersuit] doesn’t make the details clear, so he [Mr Musk] has the freedom to do whatever he wants. So the number to me, it doesn’t mean anything,” he said.

The BBC also spoke to Michael Kearney, who created another bot-measuring tool, Tweet Bot or Not, who pointed out the same thing Yang keeps trying to explain:

“Depending on how you define a bot, you could have anywhere from less than 1% to 20%,” he says.

“I think a strict definition would be a fairly low number. You have to factor in things like bot accounts that do exist, tweet at much higher volumes,” he said.

Of course, the reality remains that this case isn’t actually about bots and spam. Musk is leaning heavily on convincing his adoring fanbase it is, and as such, a tool like Botometer serves the job. He needs propaganda, not facts, and thus any tool will do, no matter how misused.

Filed Under: botometer, bots, elon musk, keicheng yang
Companies: twitter

Polling The Public About Social Media Policies Turns Up Nothing Particularly Useful

from the say-what-now? dept

Someone emailed to call my attention to some new survey results out of the University of South Florida’s Center for Cybersecurity, which contained public opinion polls about internet regulation (and gas prices, but that’s a bit outside our wheelhouse). The key part that was highlighted to me was:

More than half of Floridians (52%) say that platforms such as Twitter are “private spaces” that should be regulated only by private companies. Far less (28%) view such platforms as “public squares” where government should regulate content

And, of course, that’s interesting, because the Florida government, led by Ron DeSantis, eagerly passed a law to argue the exact opposite. Of course, that law has since been tossed out (by two separate courts, no less) as unconstitutional.

So, for all the talk of how this content moderation law was to help Floridians, it sure seems like most Floridians don’t really want DeSantis dictating social media content moderation policies either.

Of course, as you dig deeper into the data… it gets less interesting, not more.

First off, the survey design here is awful. To an embarrassing degree. The question regarding government regulation of the internet frames the question as meaning that if the government regulated social media it would do so in a manner forcing sites to remove “false, misleading, or hateful” content which is literally the exact opposite of what Florida’s social media regulation actually does. It compels sites to keep that content up.

Both of these approaches are equally unconstitutional, but it’s weird to frame the regulatory push as only going in one direction, when in the very state where this poll is taking place, the actual regulatory attempts are exactly the opposite kind of unconstitutional.

The Center also polled the Floridian public what it thought about Elon Musk’s purchase of Twitter and his ideas of what Twitter should do, and it includes the line that 50% of responses believe Twitter should “only limit offensive content if it’s illegal.” The breakdown here is also kinda weird:

Of course, that’s not so simple. Nearly all offensive content is not illegal. And… the respondents seem to recognize this, because just a few questions later they’re asked if Twitter should remove content deemed “false/misleading” and “harmful/dangerous to individuals or groups” and in both cases, respondents overwhelmingly said such content should be removed.

They were also asked if social media platforms “have a responsibility to restrict content that is false/misleading” and overwhelmingly people agreed.

But, most of that content is not illegal.

So, according to this, Twitter should only remove content that is illegal, but also not only should remove lots of perfectly legal content, but indeed has a responsibility to do so.

Perhaps this survey really says more about people’s understanding of what speech is legal than really how social media platforms should act.

For what it’s worth the survey’s strongest point of agreement seems to be that Twitter needs to work on its bot problem… and again, I don’t get that. It seems like a narrative issue, rather than reality. Elon Musk has made “bots” and “spam” a central theme of his whole takeover experience, but the vast majority of Twitter users don’t run into many bots or spam. The biggest accounts do, but most accounts don’t.

And since this question on the actual survey was framed as eliminating “non-human accounts,” that’s also weird, because many, many “non-human” accounts are actually quite useful. Some are informative — like bots that tweet earthquake reports or weather forecasts. Some are just entertaining, tweeting out random trivia or artwork. It’s the spam accounts that are a problem, but not all bots/non-human accounts are spam, and not many users really have to deal with that much spam.

So, the only thing this survey really seems to show is that you don’t learn much from polling random people about social media policies. Except maybe (1) they don’t understand how the 1st Amendment works and (2) they have been suckered by a narrative about bots.

Filed Under: bots, content moderation, elon musk, florida, free speech, regulations, survey
Companies: twitter, usf

Silly, Pandering Politicians Introduce Silly, Pandering 'Cyber Grinch' Law That Would Ban Buying Bots

from the the-grinch-turns-out-good-in-the-end dept

In December of 1983, I had just turned 9 years old, and all of my friends wanted Cabbage Patch Kids dolls. They were everywhere, and are remembered as one of the most well known holiday crazes in which scarcity of the toy, and overwhelming demand, resulted in parents absolutely losing their minds trying to find the dolls. My parents, instead, told me that the dolls were impossible to find, or super expensive if they could be found, and told me to expect something else instead. I never got a Cabbage Patch Kid, and I survived the experience (and learned a bit about supply and demand… and mass hysteria).

Anyway, I’m thinking of that experience from nearly 40 years ago today upon reading about the new “Stop Cyber Grinches from Stealing Christmas” bill, which has been announced via a press release with no actual bill text attached (which really shouldn’t ever happen). However, as described, the bill would effectively outlaw “bots” that buy up all of the popular toys in order to resell them at jacked up prices:

On Cyber Monday and with holiday shopping underway, Representative Paul Tonko (D-NY), Senator Richard Blumenthal (D-CT), Senate Majority Leader Charles E. Schumer (D-NY) and Senator Ben Ray Luj?n (D-NM) today announced the introduction of the Stopping Grinch Bots Act. Their bicameral bill will crack down on cyber Grinches using ?bot? technology to quickly buy up whole inventories of popular holiday toys and resell them to parents at higher prices. These third-party sellers use bots to bypass security measures and manipulate online sales systems to buy toys, leading to some toys being almost impossible to buy online or in stores at retail prices, exacerbating shortages caused by stressed supply chains.

?At a time when families should be able to spend time with their loved ones, digital ?Grinch bots? are forcing Americans to scour online sites in the hopes of finding an affordable gift or paying exorbitant prices for a single toy,? said Tonko. ?These bots don?t just squeeze consumers, they pose a problem for small businesses, local retailers and other entrepreneurs trying to ensure they have the best items in stock for their customers. Our Grinch Bots Act works to level the playing field and prevent scalpers from sucking hardworking parents dry this holiday season. I urge my colleagues to join me in passing this legislation immediately to stop these Grinch bots from stealing the holidays.?

And, sure, it sucks that bots are buying up popular items and then jacking up the prices, but is that really a situation that Congress needs to get involved in? Not all bad things need new regulations. And… I’m not one to necessarily say that the free market solves all problems, but this certainly seems like one that the market itself can and should deal with on its own. Companies can produce more of hot products. Retailers can put in place technical solutions to deal with bots and bulk buyers. And, like me and the Cabbage Patch Kids, kids can learn that they don’t actually need the hottest toy on the planet (and parents can realize they don’t need to buy their kids those products either).

I’m sure it must be good politics for a bunch of elected officials to claim that they’re “saving Christmas,” but that doesn’t mean that it’s good policy.

Filed Under: ben ray lujan, bots, buying, christmas, chuck schumer, demand, holidays, paul tonko, richard blumenthal, supply