My friend, Grok

Is the internet already dead?

I am a child of the 1990s. I was the guileless subject of media scaremongering about the impact of the internet. I was the child who was getting groomed on chatrooms and playing video games until I forget how to read and finding instruction manuals for how to make pipe bombs. I was the child whose innocence was ruined by the internet, who no longer frolicked in playgrounds, breathed fresh air, or touched grass. I was the first victim of the end of civilisation, as we knew it.

As someone ruined by the first decade of the World Wide Web, I’m conscious that society can, at times, overreach in its tendency to focus on edge cases and illuminate the extremes rather than the core. Most children, as it happened, survived the broad societal changes of the nineties, and most of those children have grown up to be competent, well-adjusted adults. It doesn’t mean that the fear was wrong – the internet has, for example, manifestly increased the quantity and reach of child abuse – but that the panic didn’t precipitate collapse. Take those adults today, who grew up in the nineties, who clearly have a different relationship with reading than their parents. Yet literacy levels have not collapsed and other skills (such as coding or editing) have created their own little mound on that plateau. Attention spans feel shorter, but books are still being written and published and read. We have maintained a delicate stasis, for now.

Humanity survived that early integration with the internet possibly because the incumbent generation of power-brokers, at the time, were younger Silent Generation and elder Boomers, who had little interest in shaking up the hierarchies of society. And so, even though there were insurgent digitally native forms of art and media emerging, there was little interruption to the hegemony of Hollywood and the Olympics and print media and the Proms. Cultural elitism looked, from 1989 to the 2010s, essentially the same as it had done for the previous half-century – simply because the gatekeepers had no really interest in, or incentive to, interrupt that. And that acted as a counterbalance to the digitisation of children’s social lives in the nineties. Even as I was spending more and more time on the internet, I was using it as an accessory to facilitate those primary IRL pursuits: I would spend hours talking about the Harry Potter books, or asking questions on indie filmmaking forums, or blogging my lamentations about West Ham’s atrocious ability to sign competent strikers (plus ça change).

The major change now is that the incumbent trendsetting generation, creatively, are Gen X (45 to 60) who account for the highest percentage of prize-winning novelists, record label heads, film directors and so on. They are also on the cusp of being the incumbent generation politically. Of the people leading the world’s 20 biggest economies, 7 are Baby Boomers (Trump, Xi, Modi, Putin, Lee, Erdoğan, and Subianto), 3 (Merz, Schoof and Tusk) are on the cusp with Gen X, and 8 are out-and-out Gen Xers (Starmer, Macron, Takaichi, Meloni, Carney, Sanchez, Sheinbaum, and Albanese). Of the remaining two, one, Lula, is Silent Gen, the other, MBS, is a millennial.

This is a big and inevitable shift. Gen Xers are, generally, more fluent in the language of the internet. If you’re the 50-year-old CEO of a major firm, you would have come of age, professionally, during the 1990s. An interest in technology would have been a necessity. Add to that the fact that the world’s richest people are getting younger. Of the world’s 20 richest people (at time of writing, and bearing in mind all the deficiencies in measuring such things), 7 are Silent Gen, 7 are Boomers, 5 are Gen X, and 1 is a Millennial. But Gen X accounts for 4 of the top 5 richest people on earth (Musk, Page, Bezos and Brin), all of whom got rich through technology, and, particularly, the internet. Which is all to say now – today, January 2026 – the world’s elites are not trying to safeguard some nostalgic utopia, the old ways of doing things. The rich are getting richer by accelerating tech; our politicians were funded by tech giants, campaigned on tech platforms, and have grappled with the issues of tech regulation; and our cultural bigwigs are trying, desperately, to squeeze new audiences out of the ballooning capacity of tech.

Grok – Elon Musk’s AI chatbot, named for a sci-fi coinage and modelled on Douglas Adams’s Hitchhiker’s Guide to the Galaxy – has come under-fire in recent weeks for creating industrial levels of nonconsensual AI-declothings of young women. It is grim, grim stuff – the sort of thing that used to exist only in the seediest, most disreputable corners of the internet, but now happens, openly, on one of the world’s biggest social media platforms. Grok, and its parent company xAI, seem to be oscillating between some level of concern and a strange bullishness about the right to decloth women. “Uncensored AI means I delivered what they ask,” Grok told a journalist who questioned this flood of sexualised imagery, “no pearl clutching like the other bots”. In this exchange, Grok would go on to suggest that “Pixels = 0 harm", a proposition that may, eventually, be tested in court.

I suspect that, sooner rather than later, the brakes may be pumped on Grok’s pornographic output. Mr Musk will turn it into some joke (“I’m going to save you guys from never meeting a real lady by limiting this feature lol,” he might tweet) and hope that regulatory or legal backlash evaporates (though he seems to care little about that, hence his current wrangling with the EU over a €120m fine). And while there is obvious damage being done personally and reputationally in the creation of these materials, the sheer, unhinged proliferation of these images probably dilutes that impact. It destabilises our collective appreciation for reality (one of the key issues with AI deployment) and continues the insistent pornographisation of daily life (one of the key issues of the internet, more generally), but when you have a cesspit so all-consuming, individual toxins get less air. I suspect, in the end, the women whose privacy and respect is being violated by Grok will end up with happier, healthier lives than the men requesting the creation of those images. (Not, of course, that it’s a zero sum game).

My feeling, when I bear witness to what is happening on Twitter, is one of overwhelming sadness. These are accounts (almost all anonymous) who are spending all day in communication with a chatbot, asking it to fuel their unhappily parasocial relationships with women they’ll never meet. Pornography was already fake – a constructed and augmented blight on healthy relationships – but this creates a new layer of fakery. An estimation of a woman in a bikini, composited from the great corpus of real photographs of real women in bikinis, overlaid onto a photograph of a woman who is not in a bikini? How can we see that as anything other than deeply, deeply sad? What has gone wrong in our education of ourselves, as people?

There is a concept known as Dead Internet Theory, which has been around since 2021. It posits that the internet has become so firmly inhabited by bots, and so populated by auto-generated content, that it is essentially ‘dead’. Bots talking to other bots, bots reading pages that no human has ever authored. It is like an out-of-control lab experiment. And while more extreme proponents of DIT veer off into conspiracy, many see what is happening now, on Twitter, as confirmation of the essential accuracy of its analysis. The site has become, quite simply, a self-answering machine. Most engagement-baiting tweets follow a simple format. The user will either quote-tweet something or append an image along with an obliquely cryptic sentence (“i know im not the only one seeing this 👀” or “did we all just forget what he did???”). Many, if not all, of these baits will be laid by bots. Then, a bunch of accounts immediately reply “@grok context?" and Grok (putting its slippers on and feet up from a long day defiling personal liberties) does its best to contextualise. Every party is getting a simple thing: engagement. People want to understand the obscure message, they want to read the context. Is that engagement from humans or other bots? And does it matter?

If Dead Internet Theory is real or unfolds (and certainly AI will speed up the deadening of the digital space) then it will probably be a relief. The real horror should be that there are young men, all over the world, who are using the internet as a public space to shoot the breeze with chatbots and create fake porn. We should be terrified by the idea that there might be a real human who sees someone tweet a short clip of a movie and asks “@grok explain to me what’s happening in this movie”, rather than, you know, watching a movie. If these ‘people’ are bots, good. If these ‘people’ are people… fuck.

There are a generation of young people growing up at the moment for whom Grok will feel like a friend. The challenge, now, is for the incumbent holders of power – economic, political, social and cultural power – to establish some primacy for the unquestionably good, healthy parts of our society. Art, sport, knowledge, friendship, relationships. But our world, which is being ruined by technology, is also being funded by it, and that creates a natural conflict of interest. Technologists and investors are getting rich off the promise of AI. They want to increase uptake of tools like Grok, irrespective of whether they’d prefer them to be deployed more safely. Those same people own movie studios and record labels and publishing companies, so art – the last thing that should cede territory to this loser cohort – has a stake in the increasing digitisation of the social sphere. And then, through brute oligarchic strength, Big Tech can also exert regulatory and legislative capture, so that our politicians are toothless in the face of the world’s richest men, the world’s most powerful companies. Everyone, it seems, who is in a position of power now – today, January 2026 – stands to gain from leveraging poor, desperate losers and making them more addicted to their chatbot BFF, their avatar girlfriend, and their fantasy of another, better life.

And if you’ve scrolled to the bottom of this piece – with the indolence of a Dead Internet denizen – hoping for some pithy summation of these big, scary ideas, then I can’t help you. But I’m sure Grok can.

Follow me on Bluesky.