He created Grand Theft Auto. Now he’s back with a novel about an AI that hijacks your mind – BBC

Dan Houser was one of the masterminds behind revolutionary video game series Grand Theft Auto.

Now, after leaving Rockstar Games and launching his own company, he’s released a debut novel about a very different type of game.

A Better Paradise is a dystopian vision of the near future in which an AI-led computer game goes rogue.

Set in a polarised world, it finds Mark Tyburn attempting to create a virtual haven for people to find sanctuary and reconnect within themselves against an all-consuming social media hellscape.

But it all goes wrong when it ends up unleashing a mysterious, sentient AI bot named NigelDave into society – “a hyper-intelligence built by humans” – flaws included.

Readers get to see his thought processes as he struggles with “infinite knowledge and zero wisdom”.

“What would an incredibly precocious child, who remembers everything he ever thought – because computers don’t forget things – feel like when he started talking?” Houser says.

It feels a bit like A Better Paradise predicted the future.

First released as a podcast, the book comes as AI’s continued boom means the sector’s big seven companies are now collectively worth more than China’s economy.

But Houser says he began writing the book “a good year” before OpenAI’s ChatGPT went live to the masses in 2022, complete with a logo eerily similar to his fictional creation.

Instead, it was humanity’s technological dependency during Covid – at a scale he’d underestimated – that inspired his thinking.

In his novel – which sometimes feels monologue-heavy – Houser envisions a hyper-digital, alienating world where people retreat from deepening political problems into a spiral of social media and generative AI.

Enter CEO of Tyburn Industria, Mark, who dreams of building the Ark, an immersive gaming experience users can enter in order to reconnect with themselves. It generates a world and mission tailored to each player’s innermost wants and needs.

But during testing, the Ark becomes a Pandora’s box of addiction. Some players find joy; others encounter terror. One even reconnects with his dead sister.

Meanwhile, a rogue AI bot named NigelDave slips into the real world, controlling minds and engineering realities no one can control.

Mined for advertising, people are left wondering if their thoughts are genuine. Everything is tracked, and nothing is secure. As climate emergencies intensify, society falls to pockets of civil war.

The only way to escape is to “drift”- which means hiding from a thousand algorithms by living off-grid, constantly moving and suppressing maddening paranoia that your thoughts are not your own.

To the reader, NigelDave feels like a nightmare ChatGPT gone wrong.

The AI tool recently reached 800 million weekly active users, according to boss Sam Altman, and Houser believes some people are becoming dependent on the technology’s affirming “human veneer”.

Microsoft’s head of AI Mustafa Suleyman has warned of a rise in AI psychosis – a non-clinical term describing incidents where people increasingly rely on chatbots like Claude, Grok and ChatGPT and become convinced that something imaginary has become real.

In some cases, the chatbot fuels grandiose fantasies about future opportunities. In others, it presents in a romantic connection. More troubling are reports of parents saying bots have encouraged their children to kill themselves.

In response to the increased scrutiny, ChatGPT creator OpenAI recently tightened its welfare protocols, with updates designed to ensure its chatbot responds “safely and empathetically to potential signs of delusion or mania”.

The dizzying algorithm-fuelled society in which NigelDave is unleashed mirrors ours too.

Parents worry about exposing their children to false information or harmful content. Last year national police chiefs described the “quite terrifying” misogynistic radicalisation of boys and young men. And in 2014 Facebook admitted manipulating the news feeds of nearly 700,000 users without their consent to manage the emotions they were exposed to.

“As a parent, you always worry about anything that you expose your kids to that is going to either give them false information or simply bombard them with too much information,” Houser says.

But is it bold for a video game creator to be warning of these dangers – given the long history of video games themselves being accused of making young people violent?

Houser insists there’s a difference.

“We always had the data about game violence, and it was very clear: as people played more video games, youth violence went down.

“Whatever people were claiming, we knew the opposite was true.”

Psychology professor and game violence researcher Pete Etchells says numerous studies have shown “no meaningful effect of playing violent games on aggression”.

AI models and social media are another matter – a “new paradigm” in altering behaviour that gaming never threatened, according to Matt Navarra, social media consultant and author of the Geekout Newsletter.

He says dismissing concerns as a GTA-style moral panic “understates what is changing”.

“We’re talking about external systems that can potentially shape people’s beliefs or manipulate attention, personalised experiences, nudge behaviour or even influence identity and emotional states.”

Could Houser have pushed forward with A Better Paradise at Rockstar? “I don’t know if I would have had the bandwidth to think about it,” he tells me.

He’s previously described the fatigue of managing vast open-world sandbox games like Red Dead Redemption and GTA as playing a role in his departure.

The goal with his book was to create “something truly different in this era of crazy media saturation”.

So where next? He’s already writing the second instalment of the series, and plans are under way to develop a video game, for which he promises the visuals are ground-breaking.

A key message, he says, is to not let your device – or AI – “tell you what to think”. Otherwise, Houser argues, “you’re giving up control to your phone”.

His biggest fear, as creator of worlds, is losing imagination because of the endless torrent of algorithms. Sometimes after scrolling for hours, he realises: “I’ve not had an idea all day”.

“If you go offline for a bit – sometimes I make myself go for a walk with no phone, you start to have ideas.

“A human is better thinking than not,” he says. “Thinking is a privilege.”