I’m late to the party, but I’ve just seen the viral video of the crowd at SXSW’24 booing a reel of people either promising the magic future AI will bring to our lives, or claiming it to be “without alternative”.
It’s good to see that many people do not mindlessly buy into the hype. Still, this event speaks to a deeper shift in public perception of AI. As Brian Merchant writes in his brilliant article (link here):
“For the buzziest tech of the moment to get shouted down at *SXSW* speaks volumes about the scale and nature of the animosity generative AI has amassed. The tech is seen, here, as exploitative by tastemakers and *by technologists*.”
But it’s not just the public perception that leading AI giants like OpenAI have been trying to plant in our collective understanding that is falling apart. It seems the actual narrative of AI is untangling, as people become more familiar with what it does for them and how does it make them feel. And it’s not so attractive, as it was once promised.
Because where are all the promised gains in efficiency? Where is the better world? All we see is a wasteful technology that’s propped up by VC, NVIDIA and Microsoft money that’s easily traced to see further centralisation of our digital infrastructures while trashing our digital spaces even more.
After having played with ChatGPT and other AI mainstream tools, many people don’t really integrate it into their practices because these systems and the results they spit out are just not that good. And if you care about quality, anything after quick sketches in the beginning of the process, should be labelled “believe/proceed with caution”. I sometimes wonder if those requested by me bullet points and statistically constructed mumbling and verbosity actually speeds my creative process, or I use more energy to sift through what I wanted to say than if I started with tabula rasa in the first place.
Sure, some AI systems will stick around. Especially the non-generative stuff (for image processing, object detection, bioinformatics, maintenance prediction, pattern recognition in data streams, heuristics to detect errors in work products, etc.) but a lot of the gen AI field is just spam creation. More and faster than ever. And it’s actually making some – especially creative things – harder to achieve.
Famous film critic Roger Ebert said:
“The Muse visits during the act of creation, not before.”
Roger Ebert
so maybe the booing is people coming back to realising this simple fact. You are not creative and then create something, you become creative by working on something and trusting the process. Creativity is often thought of as a spontaneous burst of inspiration or a flash of genius. Creativity is rather a byproduct of work.
In this way, AI is deeply dehumanising: making the spaces and opportunities for people to grow and become an extraordinary human, smaller and smaller. Constraining our potential and limiting our aspirations to past mediocrity, and believing it this is the best we can get.
Unless used wisely, AI may lead to Averaged Intelligence more than Augmented Intelligence, and while this may be good for quantity based work, it will render terrible results for the quality based one. AI allows below average people to level up, enabling them to be rather good, thus, reducing the ability of exceptional people to stand out, slowly diminishing discernment and critical thinking. Sure, now everyone will be able to ‘code’ or ‘create movies’, but it may become incredibly hard to find the really good, well thought-out work among the sea of average.
Over time, companies may get overwhelmed by adequate but uninspiring work. Standards can slip lower without anyone realizing it. Trivial tasks that don’t really matter become easy to generate automatically. The volume of unnecessary busy work actually grows instead of shrinking. Rather than trying to cut out the excess, people automate the tedious stuff and look the other way as meaningless churn piles up.
That is what is being booed: The salespeople of mediocrity who’ve made it their mission to promise us a personalised Neverland. The lie that only tech can and will save us. The lie that datasets skewed toward colonial, white, Western perspectives, combined with some statistical analysis, will magically forge a utopian future for all. The lie that we have no choice, no alternatives, so we better submit to it.
It becomes harder and harder to maintain the focus to be exceptional, and to go beyond plausible, likely, typical work, seems rather risky.
And while I remain largely optimistic about what AI can offer, the greatest threat from it is that we can slowly become lazy, sloppy and useless. The better it gets, the more likely we had to care less. If it could reply to 95% of your emails brilliantly….why bother checking them at all? And slowly but surely the construct of a company, the togetherness, the purpose, the sense of accomplishment dwindles. Creativity is born in chaos and a self-satisfaction comes from effort.
It’s still time to rebel against the mediocrity. But rather than applying a baseball bat against your screen, apply an ink against a paper notepad instead, and let your creativity unfold.
