The age old saying ‘If you think hiring an expert is expensive, try hiring an amateur’, seems to be starting to apply pretty well to all of this AI frenzy as well.
I’m not an AI hater, I also use it to an extent and I see it’s usefulness in speeding up certain parts of the work processes at Supersymmetric, but I can’t subscribe to the mindlessness of labeling everything ‘AI powered’ and thinking that it’s automatically better, or choose to go dumb by outsourcing any last drop of remaining intellectual effort to it.
What is happening Today seems to be the same kind of situation as the one from the beginning of the 1900’s when adding a certain known harmful substance to a wide array of over the counter remedies was pushed by the corporations that invented it as being a very safe to use universal solution to any kind of ailment.
But as the hype starts cooling off and the novelty starts wearing away, reality will slowly start to hit. And the reality of it is that it isn’t as ‘intelligent’ as the ones pushing for it want everybody to think it is, it isn’t as autonomous and it isn’t as useful on it’s own for large scale use without the oversight of someone who already knows pretty well what they’re doing.
Not to mention that it isn’t free either. The cost for business is only shifted from paying someone who knows what they’re doing, to paying hefty invoices for servers and GPU-s running energy-hungry models, that are slowly becoming worse, being contaminated by their own over digested and error prone outputs, run by the ones selling it as an universal solution to making more money. And what fun it will be in a couple of years time, when someone will have to clean up the mess created by businesses that fell for the good old ‘silicon valley group think’.
While it’s being energetically splattered on every corner of life and business, mostly without a clear purpose, it only seems to be justified by the obsession of a few in the top 1%, to make it seem like it’s ‘adding value’ in the desperation to find any viable revenue streams for it. They will continue to go out of their own way pushing for AI to increase the value of their shares in what slowly seems to become a circular economy made up of AI companies, chip manufacturers and data center builders all mangled up in a downward spiral. Instead of enabling people to think better or to solve actual real problems for a change, which was it’s original promise, the scene just seems to be flooded with vibe coding, vibe designing, and more recently generating AI err*tica, which seems to be a huge step away from the original promise, but nonetheless one of the biggest markets on the internet. In a very fickle world of too many intertwining complexities, and with the existing developments I don’t think AI will take us very far. I might be wrong, but if there’s one thing none of us can escape, that is human nature, and anything that seems to be too good to be true, usually ends up being too good to be true. Sooner or later the ‘Ponziness’ of the thing will start shine through all the manufactured ‘wonder’ and the empty promise of future riches.
And to get back to where I started from if someone thinks that hiring an expert is expensive, they should try hiring a top consulting firm using AI without proper oversight and find out, just as the Australian government recently did, paying $400.000 to a ‘top consulting firm’ and getting in exchange a report full of AI slop made up of hallucinated statistics, invented quotes, and fictitious sources. I might be blind to a lot of things and definitely not knowledgeable enough to a lot of what’s happening in AI but from where I’m standing AI still mostly seems to be just a cool party trick.
Artwork: Scholar sharpening a quill pen • Gerrit Dou