Twitter’s Banning of Journalists Could Help to Establish a New Approach to Content Moderation in the App

Between all the angst and arguments caused by Elon Musk’s most recent move to suspend journalists from the app, between all the anger baiting and trolling from both sides of the political spectrum, there may actually be some new seeds of uneasy agreements form, from which we may see a new moderation approach grow and flourish within Elon’s ‘Twitter 2.0’ experiment.

I suspect this will not be the ultimate outcome, but the latest sequence of events has shown that there is a threshold at which almost everyone agrees that there should be some moderation of the app.

We even now have a direct target for it, a barrier with which to base those future decisions.

In summary, earlier this week, Twitter suddenly banned an automated account that provides location information about where Elon’s private jet is. According to Musk, someone used this information to track him down in LA, eventually confronting a private car, which he wasn’t actually in, and harassing those inside, including Musk’s youngest child.

The rise in threat level prompted an immediate, furious response from Musk, who not only banned the @ElonJet account, but also a new twitter policy which prohibits all forms of live location tracking of individuals via tweet.

Which, as I’ve written, is likely to have implications beyond the fringe case he’s trying to deny, and may even provide a new avenue for government censorship of civil dissent.

But nevertheless, Musk implemented the new policy anyway, which he also extended the next day to anyone, including journalists, who dared to share a link to the @ElonJet account on other platforms — which, according to Musk, was done in an attempt to avoid the new rules.

Which it wasn’t, but Musk banned a handful of prominent tech reporters anyway, sparking a whole new wave of criticism about press freedom, and Musk trying to suppress dissent, using effectively authoritarian powers in his new use application.

According to Musk, the actions he took were merely an extension of his new rules, which were designed to protect his family. But for others, they are a significant excess in power, which has also raised eyebrows EU officials which have strict rules around freedom of the media built into the operating parameters of social platforms.

Still, Musk remained steadfast in his actions and even jumped into a Twitter Spaces chat with several journalists to defend his decision.

You’ll note that the @ElonJet account is a speaker in this space, which shouldn’t be possible because the account is actually suspended. Several other suspended accounts were also able to join the Space and interact as usual, a bug that Musk eventually used as an excuse to end the conversationand the entire Spaces functionality altogether, so the Twitter team could work on a fix (the Space went on for about half an hour after Musk’s brief appearance).

Musk then went back to his own safe space and exchanged tweets with the same five or so people as always who offered support and endorsement of his actions.

This is the core claim of Musk’s actions to prevent location tracking, which is not entirely correct. Musk did apply for a PIA exemption in October, which would limit public tracking of his private jet, but it’s unclear if/when that exemption took effect. And there are ways around it – basically, it’s possible to track the location of Elon’s private jet through publicly available data, while any tracking of that doesn’t extend to what happens outside of airports. The claim that this was some sort of elaborate personal tracking operation (Musk has said that these accounts publish “assassination coordinates”) is a misnomer at best.

But that’s the line Elon and Co. are taking. Musk’s friend Jason Calacanis also jumped ahead of Musk on the Twitter Spaces chat, repeatedly asking speakers “how would you feel if someone from Elon’s family got hurt” as a result of the tracking data being published.

Which is a deliberate obfuscation of the core interest, or a blatant misunderstanding. But here’s the thing – on one side of the debate you now have Elon’s biggest supporters, who are predominantly conservative commentators, saying things like:

But on the other hand, they say it’s okay for Elon to make baseless claims of pedophilia against innocent people (which has happened more than once now) and unleash his hordes of supporters on people he personally dislikes. Musk seems to believe that this type of accusation, which can also cause real harm, should be allowed, but public flight sharing should not.

Because one is a risk, and one is not?

Indeed, after being recently named and shamed by Musk about his work at Twitter, the platform’s former Trust and Security Chief Yoel Roth was forced to flee his home due to fears that Musk’s supporters might target him in real life.

Musk’s supporters expressed little sympathy for Roth’s plight.

But the same question that Calacanis posed to journalists can be posed to Musk and his fans in this case – what if Yoel Roth was actually attacked as a result of Musk’s ‘name and shame’ approach?

The question itself gets to the heart of the moderation issue at social platforms, which is not about political censorship and calls for content along ideological lines, as suggested by Musk through his recent ‘Twitter Files’ disclosure of internal discussions about important moderation issues.

If anything, social platforms want to leave as much borderline content on the platform as it generates more engagement – ​​because as previous research has shown, the closer people get to the borderline of the platform’s rules, the more interaction their posts see.

Social platforms are incentivized to leave the worst, most divisive comments — but at some point they have to ask if that comment or post could actually lead to real harm.

In this sense, both sides of the discussion here are absolutely correct – Elon Musk should not besmirch former employees by name and make them a target, as this can lead to real world damage. An account dedicated to sharing Elon Musk’s private flights, whether it’s publicly available data or not, probably shouldn’t put it on a platform of millions of people, because it could also lead to real harm if consequence of that.

This is the right threshold we should use for critical moderation decisions – which also relates to:

Elon Musk has been a vocal critic of Twitter’s previous decisions to censor and suspend the former president. But Trump’s comments, seen through the same prism of assessment, could indeed lead to real harm.

This is why this comment was removed, and what ultimately led to Trump being banned.

As Musk himself said in his brief appearance in yesterday’s Spaces chat:

“In the future there will not be any distinction between journalists and ordinary people, everyone will be treated the same. You are not special because you are a journalist, you are a Twitter user, you are a citizen.”

The same approach should apply to politicians – if this is the barrier against which we determine what should and should not be allowed in the app, all users should be held to the same standards, and the red line should be ‘potential for harm’.

Could this tweet lead to real world damage? Then it should be removed.

There is nuance and complexity to that question, as it is difficult to ascertain the actual potential for harm in each comment. But this has forever been the challenge of online content moderation, and the positive here is that we’re getting to the heart of it quickly with Elon and Co., which could lead to significant reform in approach.

Although I suspect it won’t. Right now, the platform is split into sides, with both accusing each other of opposing their perspectives, and Twitter seemingly benefiting (at least in some ways) from the increased engagement.

But in fact this incident shows that they are not in opposition. The Musk jet-tracking controversy shows that both are actually closer to agreement than they seem to realize, with the ‘free speech’ crew acknowledging that there are limits to it, and the left-leaning group conceding that, yes, there can be harm due to an account tracking the location of Musk’s vehicles.

So the meeting point here is ‘harm’ and the potential of a tweet to cause it in real life.

Now we know the limit, which seemingly everyone accepts. Hopefully, Musk and Co. can use this to formulate more effective moderation parameters moving forward.