Story audio is generated using AI
Regulation is a bogey word. We’re taught to fear it in most liberal democracies. It summons images of the left-brained, buttoned-up bureaucrat and the persnickety policy that limits the flourish of a free-market economy. Yet all but the most hardened anarchist would not dispute that it is necessary on some level. Humanity is currently grappling with understanding, let alone controlling, forces of a truly existential nature — the technological disruptions that are influencing every facet of society: social media, online gambling, gig work and — the big one that forces its way into every conversation — artificial intelligence (AI).These topics share two key similarities. They have developed exponentially in the last decade and a half, and it has proven impossible to map out their trajectory in any meaningful way. The global legal scaffolding has struggled to keep pace with soaring technological innovation, which often leaves us scrambling to pick up the pieces after the damage has been done. Last week Meta had to cough up $18bn to settle a social media addiction trial. It has also agreed to implement strong child-safety protocols on its platforms. It’s a seminal moment for the company and its peers. For most of its existence Meta has sought to sidestep responsibility for what goes on its pages, with the argument that it is a platform, not a publisher. It does not produce content; it only functions as a global town hall.No one is buying that anymore. The supremacy of the algorithm — and its innumerable effects — is irrefutable. The US settlement follows concerted legal debate on social media in the UK and Australia. The latter in particular has paved the way for thinking around identity verification and age restrictions. This trend seems only logical. A far more interesting question to ponder is whether we could have got here a lot sooner.My generation was Facebook patient zero. Many millennials were in high school or university when the website went global. We huddled together in computer labs, ogling one another’s “walls”. It was mostly clean, innocuous fun. Those a little younger than us faced an altogether different prospect: an entire adolescence stalked by the pressures of maintaining a vibrant digital presence. We now know, thanks to an enormous body of research, how damaging that has been.No one could have predicted the specific, all-consuming form social media would morph into. But is that enough to exonerate us for not acting sooner when it became empirically obvious that the fabric of society was being rewoven in real time? (We haven’t even begun to discuss how new platforms have invariably altered the media landscape).Our tussle with social media portends an awkward time ahead. As rapidly as that technology grew in the last decade, it is a comparative laggard to those of this one. Gambling has been around for at least five millennia. But when its addictive properties were harnessed through apps, push notifications and quick deposits, it became something altogether different. Most governments are still trying to figure out what exactly that is, and where it is appropriate for them to step in.The threat is that by the time they have answers, change may have made them redundant. That unique form of paralysis is what characterises the discourse around AI — the dominant issue of our times. It is a technology that no one denies will profoundly change our way of life, but we are nowhere near consensus on where the guardrails must be erected. Or even if they should be.Complicating an intractable problem even further is that it is often viewed through the prism of an arms race. US President Donald Trump has in fact gone the opposite way of regulation, forbidding states from doing so themselves. “We are beating all countries at this point in the race,” he wrote ahead of an executive order in December, but if prohibitive policies are put in place, ”AI will be destroyed in its infancy!”Last week may come to be seen as a turning point in how we perceive the issue, though. A remarkable open letter signed by a disparate set of major companies — from Uber and Microsoft to PayPal and Perplexity — called for “collective action on cyberdefence”. It followed the OpenAI investigation into how the Hugging Face hack was perpetrated: the July sci-fi fever dream turned real in which AI agents mobilised into a “swarm” to break free from their parameters.The letter stops short of using the “R” word, but it does carry a hefty warning for governments to shore up their cybersecurity, prioritise risk assessment and share intelligence. Whether that translates into regulation — whether national, multilateral or even self-imposed — will, for better or worse, define our future.• Feltham is Business Day editor-in-chief.Business Day
