“Best humans still outperform”: One turning point in the history of cope around artificial intelligence

In a widely shared essay, writer Oliver Sourbut dissects a pivotal moment in public AI discourse: a 2023 academic journal headline declaring 'Best humans still outperform artificial intelligence in a creative divergent thinking task.' Sourbut argues this seemingly mundane headline revealed a profound cultural shift. The very need to assert human superiority signaled that society had viscerally accepted machines as genuine competitors. This was no longer science fiction; the 'machines were coming' had become a mainstream reality check, even as tools like ChatGPT and Stable Diffusion were less than a year old.

Sourbut credits this 'ChatGPT moment' with democratizing the AI conversation. Instead of being confined to a 'plucky band of vaguely foresighted sci-fi weirdos,' the debate now involves global stakeholders. Lawyers, philosophers, engineers, economists, and policymakers are all engaging with AI's thickly textured problems—from ethics and safety to economics and national security. This broadened engagement is crucial, as Sourbut contends the challenges are too complex for any single group to solve.

The essay suggests we've moved into a new, more precarious phase. By 2026, AI is central to 'national security implications and dirty political manoeuvring,' with 'powerful and not altogether friendly forces' wrestling for control. The training wheels are off. While this introduces significant risks, the widespread awareness also creates an opportunity. Humanity, now collectively acting as stakeholders, has the chance to steer this transformative technology toward positive outcomes like world peace and unlimited bounty, rather than catastrophic ones.