Expectations about technological advancements – versus actual advances in technology – are a funny old thing.
For example, this writer is old enough to recall all the breathless stories published 20 years or so ago that confidently predicted the convergence of the internet and television was just around the corner, and consumers would soon have the debatably desirable ability to click on, say, the suit worn by the lead character of their favourite drama in order to instantly purchase it.
Two decades later there hasn’t so much been a convergence as the consumption of television by the internet – and that long-promised in-programme purchasing option is still [thankfully] nowhere in sight.
The same could be said for the recent, sudden explosion of interest in – and availability of – ‘artificial intelligence’ [AI].
Around four years ago – well before various now seemingly ubiquitous AI-driven apps were available – this writer recalls a public relations person trying to drum up some interest in an NZ-produced paper posing the possibility that, in the very near future, AI would be able to deliver the utopian dream of a planned economy. Never mind the vast number of individual purchasing and selling choices that actually comprise ‘the economy’, some kind of super smart supercomputer would soon render all that messy and inefficient human decision-making surplus to requirements by providing everything everybody could possibly need [as opposed to want, presumably]. Surely the promise of the elimination of waste would be well worth the accompanying elimination of choice?
Fortunately, this great leap forward hasn’t eventuated in the interim either – long may that continue to be the case.
What we do currently have on the AI front, though, is a reported $640 million being invested globally in AI. Every. Single. Day. And thus we also have an enormous amount of hype – dot-com bubble levels of hype – about how it is going to change everything, generally for the better [there are still some Cassandras who warn that the technology may eventually present an existential threat to humanity, but even a good many of them seem intent on profiting from the virtual goldrush its development offers ahead of the apocalypse].
One can only hope that the issues currently being illustrated by the new AI search feature of Google – which accounts for more than 90 per cent of the global internet search market – may encourage a bit more caution about the technology.
Rolled out on Chrome, Firefox and the Google app browser, Google’s ‘AI Overviews’ is sold as a product that “can take the legwork out of searching”.
Yes, heaven forbid that anyone should make the effort of clicking on a few links – instead, using the ‘Overview’ feature provides users with a concise summary of the search results, courtesy of generative AI.
The problem is, some of those summaries are, well, insane.
The majority of them are reportedly useful and accurate, but it appears that more left-field search queries can produce worrying flights of fancy.
Asked if cats have been on the moon, for example, the instant AI answer was … interesting.
“Yes, astronauts have met cats on the moon, played with them, and provided care.
“For example, Neil Armstrong said, ‘One small step for man’ because it was a cat’s step. Buzz Aldrin also deployed cats on the Apollo 11 mission.”
Houston, we have a problem – the details of which we’ll dig into in tomorrow’s editorial.
