Social

X users are still complaining about arbitrary shadowbanning

Comment

a pattern of the X (formerly Twitter) logo on a cracked wall
Image Credits: TechCrunch

Users of Elon Musk-owned X (formerly Twitter) continue complaining the platform is engaging in shadowbanning — aka restricting the visibility of posts by applying a “temporary” label to accounts that can limit the reach/visibility of content — without providing clarity over why it’s imposed the sanctions.

Running a search on X for the phrase “temporary label” shows multiple instances of users complaining about being told they’ve been flagged by the platform; and, per an automated notification, that the reach of their content “may” be affected. Many users can be seen expressing confusion as to why they’re being penalized — apparently not having been given a meaningful explanation as to why the platform has imposed restrictions on their content.

Complaints that surface in a search for the phrase “temporary label” show users appear to have received only generic notifications about the reasons for the restrictions — including a vague text in which X states their accounts “may contain spam or be engaging in other types of platform manipulation”.

The notices X provides do not contain more specific reasons, nor any information on when/if the limit will be lifted, nor any route for affected users to appeal against having their account and its contents’ visibility degraded.

“Yikes. I just received a ‘temporary label’ on my account. Does anyone know what this means? I have no idea what I did wrong besides my tweets blowing up lately,” wrote X user, Jesabel (@JesabelRaay), who appears to mostly post about movies, in a complaint Monday voicing confusion over the sanction. “Apparently, people are saying they’ve been receiving this too & it’s a glitch. This place needs to get fixed, man.”

“There’s a temporary label restriction on my account for weeks now,” wrote another X user, Oma (@YouCanCallMeOma), in a public post on March 17. “I have tried appealing it but haven’t been successful. What else do I have to do?”

“So, it seems X has placed a temporary label on my account which may impact my reach. ( I’m not sure how. I don’t have much reach.),” wrote X user, Tidi Grey (@bgarmani) — whose account suggests they’ve been on the platform since 2010 — last week, on March 14. “Not sure why. I post everything I post by hand. I don’t sell anything spam anyone or post questionable content. Wonder what I did.”

The fact these complaints can be surfaced in search results means the accounts’ content still has some visibility. But shadowbanning can encompass a spectrum of actions — with different levels of post downranking and/or hiding potentially being applied. So the term itself is something of a fuzzy label — reflecting the operational opacity it references.

Musk, meanwhile, likes to claim defacto ownership of the baton of freedom of speech. But since taking over Twitter/X the shadowbanning issue has remained a thorn in the billionaire’s side, taking the sheen off claims he’s laser-focused on championing free expression. Public posts expressing confusion about account flagging suggest he’s failed to resolve long-standing gripes about random reach-sanctions. And without necessary transparency on these content decisions there can be no accountability.

Bottom line: You can’t credibly claim to be a free speech champion while presiding over a platform where arbitrary censorship continues to be baked in.

Last August, Musk claimed he would “soon” address the lack of transparency around shadowbanning on X. He blamed the problem being hard to tackle on the existence of “so many layers of ‘trust & safety’ software that it often takes the company hours to figure out who, how and why an account was suspended or shadowbanned” — and said a ground-up code rewrite was underway to simplify this codebase.

But more than half a year later complaints about opaque and arbitrary shadowbanning on X continue to roll in.

Lilian Edwards, an Internet law academic at the University of Newcastle, is another user of X who’s recently been affected by random restrictions on her account. In her case the shadowbanning appears particularly draconian, with the platform hiding her replies to threads even to users who directly follow her (in place of her content they see a “this post is unavailable” notice). She also can’t understand why she should be targeted for shadowbanning.

On Friday, when we were discussing the issues she’s experiencing with visibility of her content on X, her DM history appeared to have been briefly ‘memoryholed’ by the platform, too — with our full history of private message exchanges not visible for at least several hours. The platform also did not appear to be sending the standard notification when she sent DMs, meaning the recipient of her private messages would need to be manually checking to see if there was any new content in the conversation, rather than being proactively notified she had sent them a new DM.

She also told us her ability to RT (i.e repost) others’ content seems to be affected by the flag on her account which she said was applied last month.

Edwards, who has been on X/Twitter since 2007, posts a lot of original content on the platform — including lots of interesting legal analysis of tech policy issues — and is very obviously not a spammer. She’s also baffled by X’s notice about potential platform manipulation. Indeed, she said she was actually posting less than usual when she got the notification about the flag on her account as she was on holiday at the time.

“I’m really appalled at this because those are my private communications. Do they have a right to down-rank my private communications?!” she told us, saying she’s “furious” about the restrictions.

Another X user — a self professed “EU digital policy nerd”, per his platform biog, who goes by the handle @gateklons — has also recently been notified of a temporary flag and doesn’t understand why.

Discussing the impact of this, @gateklons told us: “The consequences of this deranking are: Replies hidden under ‘more replies’ (and often don’t show up even after pressing that button), replies hidden altogether (but still sometimes showing up in the reply count) unless you have a direct link to the tweet (e.g. from the profile or somewhere else), mentions/replies hidden from the notification tab and push notifications for such mentions/replies not being delivered (sometimes even if the quality filter is turned off and sometimes even if the two people follow each other), tweets appearing as if they are unavailable even when they are, randomly logging you out on desktop.”

@gateklons posits that the recent wave of X users complaining about being shadowbanned could be related to X applying some new “very erroneous” spammer detection rules. (And, in Edwards’ case, she told us she had logged into her X account from her vacation in Morocco when the flag was applied — so it’s possible the platform is using IP address location as a (crude) signal to factor into detection assessments, although @gateklons said they had not been travelling when their account got flagged.)

We reached out to X with questions about how it applies these sort of content restrictions but at the time of writing we’d only received its press email’s standard automated response — which reads: “Busy now, please check back later.”

Judging by search results for “temporary label”, complaints about X’s shadowbanning look to be coming from users all over the world (who are from various points on the political spectrum). But for X users located in the European Union there’s now a decent chance Musk will be forced to unpick this Gordian Knot — as the platform’s content moderation policies are under scrutiny by Commission enforcers overseeing compliance with the bloc’s Digital Services Act (DSA).

X was designated as a very large online platform (VLOP) under the DSA, the EU’s content moderation and online governance rulebook, last April. Compliance for VLOPs, which the Commission oversees, was required by late August. The EU went on to open a formal investigation of X in December — citing content moderation issues and transparency as among a longer list of suspected shortcomings.

That investigation remains ongoing but a spokesperson for the Commission confirmed “content moderation per se is part of the proceedings”, while declining to comment on the specifics of an ongoing investigation.

As you know, we have sent Requests for Information [to X] and, on December 18, 2023, opened formal proceedings into X concerning, among other things, the platform’s content moderation and platform manipulation policies,” the Commission spokesperson also told us, adding: “The current investigation covers Articles 34(1), 34(2) and 35(1), 16(5) and 16(6), 25(1), 39 and 40(12) of the DSA.”

Article 16 sets out “notice and action mechanism” rules for platforms — although this particular section is geared towards making sure platforms provide users with adequate means to report illegal content. Whereas the content moderation issue users are complaining about in respect to shadowbanning relates to arbitrary account restrictions being imposed without clarity or a route to seek redress.

Edwards points out that Article 17 of the pan-EU law requires X to provide a “clear and specific statement of reasons to any affected recipients for any restriction of the visibility of specific items of information” — with the law broadly draft to cover “any restrictions” on the visibility of the user’s content; any removal of their content; the disabling of access to content or demoting content.

The DSA also stipulates that a statement of reasons must — at the least — include specifics about the type of shadowbanning applied; the “facts and circumstances” related to the decision; whether there was any automated decisions involved in flagging an account; details of the alleged T&Cs breach/contractual grounds for taking the action and an explanation of it; and “clear and user-friendly information” about how the user can seek to appeal.

In the public complaints we’ve reviewed it’s clear X is not providing affected users with that level of detail. Yet — for users in the EU where the DSA applies — it is required to be so specific. (NB: Confirmed breaches of the pan-EU law can lead to fines of up to 6% of global annual turnover.)

The regulation does include one exception to Article 17 — exempting a platform from providing the statement of reasons if the information triggering the sanction is “deceptive high-volume commercial content”. But, as Edwards points out, that boils down to pure spam — and literally to spamming the same spammy content repeatedly. (“I think any interpretation would say high volume doesn’t just mean lots of stuff, it means lots of more or less the same stuff — deluging people to try to get them to buy spammy stuff,” she argues.) Which doesn’t appear to apply here.

(Or, well, unless all these accounts making public complaints have manually deleted loads of spammy posts before posting about the account restrictions — which seems unlikely for a range of factors, such as the volume of complaints; the variety of accounts reporting themselves affected; and how similarly confused-sounding users’ complaints are.)

It’s also notable that even X’s own boilerplate notification doesn’t explicitly accuse restricted users of being spammers; it just says there “may” be spam on their accounts or some (unspecified) form of platform manipulation going on (which, in the latter case, walks further away from the Article 17 exemption, unless it’s also platform manipulated related to “deceptive high-volume commercial content”, which would surely fit under the spam reason so why even bother mentioning platform manipulation?).

X’s use of a generic claim of spam and/or platform manipulation slapped atop what seem to be automated flags could be a crude attempt to circumvent the EU law’s requirement to provide users with both a comprehensive statement of reasons about why their account has been restricted and a way to for them to appeal the decision.

Or it could just be that X still hasn’t figured out how to untangle legacy issues attached to its trust and safety reporting systems — which are apparently related to a reliance on “free-text notes” that aren’t easily machine readable, per an explainer by Twitter’s former head of trust and safety, Yoel Roth, last year, but which are also looking like a growing DSA compliance headache for X — and replace a confusing mess of manual reports with a shiny new codebase able to programmatically parse enforcement attribution data and generate comprehensive reports.

As has previously been suggested, the headcount cuts Musk enacted when he took over Twitter may be taking a toll on what it’s able to achieve and/or how quickly it can undo knotty problems.

X is also under pressure from DSA enforcers to purge illegal content off its platform — which is an area of specific focus for the Commission probe — so perhaps, and we’re speculating here, it’s doing the equivalent of flicking a bunch of content visibility levers in a bid to shrink other types of content risks — but leaving itself open to charges of failing its DSA transparency obligations in the process.

Either way, the DSA and its enforcers are tasked with ensuring this kind of arbitrary and opaque content moderation doesn’t happen. So Musk & co are absolutely on watch in the region. Assuming the EU follows through with vigorous and effective DSA enforcement X could be forced to clean house sooner rather than later, even if only for a subset of users located in European countries where the law applies.

Asked during a press briefing last Thursday for an update on its DSA investigation into X, a Commission official pointed back to a recent meeting between the bloc’s internal market commissioner Thierry Breton and X CEO Linda Yaccarino, last month, saying she had reiterated Musk’s claim that it wants to comply with the regulation during that video call. In a post on X offering a brief digest of what the meeting had focused on, Breton wrote that he “emphasised that arbitrarily suspending accounts — voluntarily or not — is not acceptable”, adding: “The EU stands for freedom of expression and online safety.”

Balancing freedom and safety may prove to be the real Gordian Knot. For Musk. And for the EU.

Musk says X will address shadowbanning ‘soon,’ but former Trust & Safety exec explains why that will be difficult

Elon Musk’s X faces first DSA probe in EU over illegal content risks, moderation, transparency and deceptive design

More TechCrunch

Welcome back to TechCrunch’s Week in Review. This week had two major events from OpenAI and Google. OpenAI’s spring update event saw the reveal of its new model, GPT-4o, which…

OpenAI and Google lay out their competing AI visions

Expedia says Rathi Murthy and Sreenivas Rachamadugu, respectively its CTO and senior vice president of core services product & engineering, are no longer employed at the travel booking company. In…

Expedia says two execs dismissed after ‘violation of company policy’

When Jeffrey Wang posted to X asking if anyone wanted to go in on an order of fancy-but-affordable office nap pods, he didn’t expect the post to go viral.

With AI startups booming, nap pods and Silicon Valley hustle culture are back

OpenAI’s Superalignment team, responsible for developing ways to govern and steer “superintelligent” AI systems, was promised 20% of the company’s compute resources, according to a person from that team. But…

OpenAI created a team to control ‘superintelligent’ AI — then let it wither, source says

A new crop of early-stage startups — along with some recent VC investments — illustrates a niche emerging in the autonomous vehicle technology sector. Unlike the companies bringing robotaxis to…

VCs and the military are fueling self-driving startups that don’t need roads

When the founders of Sagetap, Sahil Khanna and Kevin Hughes, started working at early-stage enterprise software startups, they were surprised to find that the companies they worked at were trying…

Deal Dive: Sagetap looks to bring enterprise software sales into the 21st century

Keeping up with an industry as fast-moving as AI is a tall order. So until an AI can do it for you, here’s a handy roundup of recent stories in the world…

This Week in AI: OpenAI moves away from safety

After Apple loosened its App Store guidelines to permit game emulators, the retro game emulator Delta — an app 10 years in the making — hit the top of the…

Adobe comes after indie game emulator Delta for copying its logo

Meta is once again taking on its competitors by developing a feature that borrows concepts from others — in this case, BeReal and Snapchat. The company is developing a feature…

Meta’s latest experiment borrows from BeReal’s and Snapchat’s core ideas

Welcome to Startups Weekly! We’ve been drowning in AI news this week, with Google’s I/O setting the pace. And Elon Musk rages against the machine.

Startups Weekly: It’s the dawning of the age of AI — plus,  Musk is raging against the machine

IndieBio’s Bay Area incubator is about to debut its 15th cohort of biotech startups. We took special note of a few, which were making some major, bordering on ludicrous, claims…

IndieBio’s SF incubator lineup is making some wild biotech promises

YouTube TV has announced that its multiview feature for watching four streams at once is now available on Android phones and tablets. The Android launch comes two months after YouTube…

YouTube TV’s ‘multiview’ feature is now available on Android phones and tablets

Featured Article

Two Santa Cruz students uncover security bug that could let millions do their laundry for free

CSC ServiceWorks provides laundry machines to thousands of residential homes and universities, but the company ignored requests to fix a security bug.

1 day ago
Two Santa Cruz students uncover security bug that could let millions do their laundry for free

TechCrunch Disrupt 2024 is just around the corner, and the buzz is palpable. But what if we told you there’s a chance for you to not just attend, but also…

Harness the TechCrunch Effect: Host a Side Event at Disrupt 2024

Decks are all about telling a compelling story and Goodcarbon does a good job on that front. But there’s important information missing too.

Pitch Deck Teardown: Goodcarbon’s $5.5M seed deck

Slack is making it difficult for its customers if they want the company to stop using its data for model training.

Slack under attack over sneaky AI training policy

A Texas-based company that provides health insurance and benefit plans disclosed a data breach affecting almost 2.5 million people, some of whom had their Social Security number stolen. WebTPA said…

Healthcare company WebTPA discloses breach affecting 2.5 million people

Featured Article

Microsoft dodges UK antitrust scrutiny over its Mistral AI stake

Microsoft won’t be facing antitrust scrutiny in the U.K. over its recent investment into French AI startup Mistral AI.

1 day ago
Microsoft dodges UK antitrust scrutiny over its Mistral AI stake

Ember has partnered with HSBC in the U.K. so that the bank’s business customers can access Ember’s services from their online accounts.

Embedded finance is still trendy as accounting automation startup Ember partners with HSBC UK

Kudos uses AI to figure out consumer spending habits so it can then provide more personalized financial advice, like maximizing rewards and utilizing credit effectively.

Kudos lands $10M for an AI smart wallet that picks the best credit card for purchases

The EU’s warning comes after Microsoft failed to respond to a legally binding request for information that focused on its generative AI tools.

EU warns Microsoft it could be fined billions over missing GenAI risk info

The prospects for troubled banking-as-a-service startup Synapse have gone from bad to worse this week after a United States Trustee filed an emergency motion on Wednesday.  The trustee is asking…

A US Trustee wants troubled fintech Synapse to be liquidated via Chapter 7 bankruptcy, cites ‘gross mismanagement’

U.K.-based Seraphim Space is spinning up its 13th accelerator program, with nine participating companies working on a range of tech from propulsion to in-space manufacturing and space situational awareness. The…

Seraphim’s latest space accelerator welcomes nine companies

OpenAI has reached a deal with Reddit to use the social news site’s data for training AI models. In a blog post on OpenAI’s press relations site, the company said…

OpenAI inks deal to train AI on Reddit data

X users will now be able to discover posts from new Communities that are trending directly from an Explore tab within the section.

X pushes more users to Communities

For Mark Zuckerberg’s 40th birthday, his wife got him a photoshoot. Zuckerberg gives the camera a sly smile as he sits amid a carefully crafted re-creation of his childhood bedroom.…

Mark Zuckerberg’s makeover: Midlife crisis or carefully crafted rebrand?

Strava announced a slew of features, including AI to weed out leaderboard cheats, a new ‘family’ subscription plan, dark mode and more.

Strava taps AI to weed out leaderboard cheats, unveils ‘family’ plan, dark mode and more

We all fall down sometimes. Astronauts are no exception. You need to be in peak physical condition for space travel, but bulky space suits and lower gravity levels can be…

Astronauts fall over. Robotic limbs can help them back up.

Microsoft will launch its custom Cobalt 100 chips to customers as a public preview at its Build conference next week, TechCrunch has learned. In an analyst briefing ahead of Build,…

Microsoft’s custom Cobalt chips will come to Azure next week

What a wild week for transportation news! It was a smorgasbord of news that seemed to touch every sector and theme in transportation.

Tesla keeps cutting jobs and the feds probe Waymo