Freedom of speech, social media & Twitter
A couple of days ago, I was watching a speech Sacha Baron Cohen made at an ADL conference (the Anti-Defamation League, a US organization created by Jewish Americans to preserve the rights of all minorities) and in which, uncharacteristically, he was talking as himself. Interestingly, that video was suggested to me on TikTok, the new Chinese-born social media giant currently taking over the world, to Americans’ manifest dismay.
Back on topic: as himself, Cohen very eloquently put forth the argument that social media is in essence a threat to our democracies insofar as half a dozen tech billionaires effectively control these few giants most of our digital communications go through today (the speech was made pre-TikTok FYI). And their business model is based on algorithms that serve more of what you want, including if what you want is xenophobic, racist, sexist and/or antisemitic content. In the name of freedom of speech, as it happens.
The comedian is far from the only one promoting this argument: many other public figures, media professionals and people at the bar down the road have said as much. And there obviously is merit to the notion that we have to control what intelligent systems offer when it comes to ideas and ideologies. But is that really the core issue?
Blaming AI
In the past few years, even more so recently with the rise of ChatGPT, blaming AI has been en vogue. With good reason: it sounds powerful, it is largely misunderstood and tech moguls talk about it all the time, starting with Elon Musk (alternatively promoting it and warning about its risks, because of course). The truth is that AI, no matter how deep, intelligent, machine-learning-enabled, is merely a tool at one’s disposal. A powerful one at that, but nothing more — meaning that it does not create suggestions out of thin air on your social media platform of choice, they tend to match users’ choices — and are based on content provided by other users, be they private citizens or organizations. Blaming AI today is like blaming TV yesterday or the newspaper before that: it as blanket blame.
Indeed, some media platforms are blatantly promoting radical ideologies. In digital media, you have Breitbart; on TV, you obviously have Fox News; in social media, you have the ironically named “Truth Social” that Trump created for himself after he got kicked out of Twitter (and refused to get back on once Musk reinstated his account). These platforms are clearly biased and every user knows it and consumes content there precisely because of that. The Facebooks, Youtubes, TikToks and Twitters of this world, however, are mainstream. In other words, they don’t specifically appeal to a target community, or at least to a much larger one (TikTok is for the younger demo, Facebook now for the older, that kind of thing). They are agnostic as in they are not professing to follow any political current or ideology, because that would quite directly diminish their user base.
In this biased / agnostic landscape, the impact of AI effectively depends on the contents it has at its disposal. On Truth Social, it will most likely give you pretty biased content no matter what you type, given that most content on the platform points towards that. On Instagram, it will depend on your tastes: could end up falling into the same category as Truth Social (albeit perhaps with better spelling) if you’re into that, or it could placate a Biden/Harris lovefest. In other words, AI as such is itself agnostic — it will work with what you give it.
The specific argument that AI enhances fringe groups on mainstream social media is also to be taken with a grain of salt. As it happens, most contents on Facebook are not far right (or far left): the majority of it is pretty lame (same argument would work for other mainstream platforms to be clear: this is not only because Facebook now caters to seniors). That being said, a case could be made that white supremacist / white nationalist groups did have pretty good success using these platforms to promote their “ideas”. Technically speaking, this mostly points to the fact that the mainstream / liberal / moderate user base is not as good at this game as it should be (see: the 2016 digital presidential war). We all have the same tools…
The final argument that AI secretly favors fringe groups ironically largely results in a conspiracy theory about conspiracy theorists. There is no evidence whatsoever that far right (or left) movements are being disproportionately promoted based on the nature of their contents — on mainstream platforms, that is. Indeed, Youtube & co are not in the habit of pushing things on you that you dislike instead of things that you do like: first and foremost, they want you to keep clicking. If you are pro-choice, you won’t be seeing a Southern pro-life preacher giving a sermon anytime soon on any mainstream platform. Unless it comes with disparaging commentary.
Enter Twitter 2.0
The real game changer today, if there is one, is called Twitter. Or rather Twitter 2.0. Or rather Elon Musk’s Twitter. What happened with Musk’s takeover of one of the largest social media platforms out there was that the waters definitely got muddied between mainstream and fringe. And that is worrisome.
For those who didn’t follow this little tech soap opera, let’s recap:
for many years, Twitter was the smallest out of the mainstream social media platforms, I would argue because its product was never quite fully-formed. Nevertheless, it boasted hundreds of millions of users across the globe with a particular proximity with (regular) media, and more recently a blockchain trope to boot;
because the company was comparatively small, its finances — and overall governance — were more shaky than others. That and the fact that founder Jack Dorsey had a habit of coming in and out the door. Either way, the business got into trouble as the digital media wars grew tougher;
Elon Musk, a longtime Twitter power-user, started showing interest in buying the platform. It made sense because he was wealthy enough to do so (thanks to the skyrocketing rise of Tesla) and because he quite visibly loved the product;
The problem, however, is that Musk is not exactly agnostic when it comes to his posts. During Covid, he actively spread anti-mask rumors, he is a very public Trump (and Kanye) supporter, his libertarian views often end up promoting highly debatable (and debated) theories… Which led the board to shoot him down.
Musk eventually offered a deal Twitter couldn’t refuse last year, and swiftly started making changes, including firing lots of people, changing some of the company’s most prominent policies, such as reinstating the aforementioned Trump account (as well as Kanye’s, which got shut down again days later when he posted more antisemitic content) and making the blue check mark a paid option, leading the New York Times to losing theirs…
Beyond the fun memes and anecdotal aspect of some of these episodes, what is unfolding before our eyes is actually quite serious, significantly more so than any AI-rising-kind of threat. A man with demonstrable fringe opinions is now in control of one of the biggest, most mainstream social media platforms on the planet. And he could decide to tweak it to actually promote extremist opinions, far right content and so forth. The reason for this is — he’s not (only) after money here. He’s after power.
And, unsurprisingly given his public statements, the kind of power he pursues is one borne out of unrest. The destructive power of fringe individuals trying to shake the system to their advantage through gaslighting, shock value statements and/or fake news whenever appropriate. To make a buck along the way, sure, but also and more critically to shape it in a way that fits their worldview better.
Put simply, Elon Musk is a way more credible threat to freedom of speech — regardless what he professes — than any robot will ever be. Perfectly summarizing this is his actual involvement in AI: a co-founder of OpenAI, which happened to create ChatGPT, he recently came out in favor of a “halt” in deep AI research. In other words, do one thing, then say the opposite and keep going while everybody is left wondering what just happened…
Now what?
To be clear, I don’t think Elon Musk, no matter how smart, wealthy or now socially connected he may be, will be able to actually shake democracy to its core, despite what Cohen and others would claim. As it happens, moderate, liberal, tech-savvy people are also on social media, fighting the good fight; and the younger users among us could detect a deep fake with their eyes closed. For the truth is that, in the long run, as was the case with the newspaper or TV before it, the internet in general and social media in particular are incredible communication tools that bring more positive than negative into the world, all things considered.
In the shorter term, though, if we want to curtail the most obvious threat that social media poses to democracy, we have to see what we can do to hamper Musk’s actions. Challenge the legality of some Twitter decisions or regulations, challenge the man’s actions directly, boycott the platform… The recent Doge-gate could be just that: the man is actually being investigated for shady dealings with the currency, so he puts the Doge dog’s picture on his website to mislead people — and mess with google searches…
There are options at our disposal, as always. And, as always, the overwhelming number of well-meaning men and women can overtake any unhinged mogul(s).