The Inevitability of AI
This essay is part of a thread on AI, data sovereignty, and Pasifika in technology. See the Pacific AI & Data Sovereignty hub for the full collection. Related: AI Is a Conversation, Getting Paid For Your Data Sounds Great.
Talofa reader,
The argument that "AI is inevitable" has been accepted by tech and non-tech professionals alike, largely with muted reservation rather than loud proclamation. I've led workshops on generative AI, and the feeling from participants is that AI is a foregone conclusion:
"If you don't keep up, you'll be left behind."
They're not saying it because they're convinced about the technology, but because of how it feels to be surrounded by it.
And that's the problem. This is a claim that needs to be scrutinised, because you may have more optionality than you're led to believe.
Leon Furze writes about the difference between "unavoidable" and "inevitable."
Unavoidable is just an apt description of the state of the industry right now—you can't turn anywhere without running into an AI-enabled or AI-powered anything. To deny that is to have your head in the sand.
But "inevitable" is different.
To say AI is inevitable is to probably have an agenda or vested interest in making the AI future a certainty.
The Audrey Watters quote sums up my take:
"Anytime someone maintains that a technology is inevitable, they're asking us to give up any say we might have over that technology—asking us to abandon our agency, the control of our future." -- Audrey Watters
What slipstreams into the acceptance of AI as unavoidable is the acceptance of whatever version and roadmap for AI the current powers dictate. And I would want to push back on that as much as possible.
I've been a cynical observer of the tech marketing and hype machine for decades now. It's hard not to when you've been face to face with the reality behind the curtain—not just the technology, but the engineers, managers and executives behind it. The reality is far from the polished marketing bullshit.
If I'd listened to the doomers a couple of years ago, I'd be behind right now in understanding what the tech can really do—both positively and negatively. If I'd listened to the hype merchants and grifters, I'd have a lot of egg on my face walking back outlandish claims about eliminating hallucinations.
And if you just took your news from mainstream tech media, you might believe there's only one playbook. But if you're tuned into the movements of tech players big and small, you'd see there are more power moves in the game than just the Silicon Valley techbro narrative. DeepSeek R1 seemingly "coming out of nowhere" and completely rewriting the narrative on model training costs—while being transparent with models showing their chain-of-thought—is a good example.
Before I go further on the anti-AI crowd: look, I get it. I hate the grifters and techbros as much as the next anarchist.
I agree—the copyright issue, stealing artists' works and training models on them, then allowing the internet to plagiarise at scale while only compensating the AI model owners. I know the models hallucinate, so why place these systems into important decision-making positions in our public sector, schools and universities? Climate change is real, and the reports of data centres' impacts on water use and the environment can't be taken lightly.
So yes, there's a lot to be anti-AI about.
But I find just being snarky and "I told you so" when one of your "AI is a bubble" predictions finally comes in—whilst entertaining, sure, but how the fuck does that help any of us? Maybe that’s the point, it’s not.
But I digress!
We don't all have the luxury of living off newsletter subscriptions. We have to contend with AI's place in the real world. Some helpful sense of organising or community to counter things in real terms would be useful.
But I'm a practitioner. I learn by doing. I get hands-on, read the papers, talk to other folks about the tech. I get a taste of it myself so I know what I'm talking about.
And like most things, the best answer lies somewhere in the middle.
You can't speak on what you haven't experienced—or at least, your words won't hold as much weight as someone who has tried things out themselves.
As an intelligent*, free-thinking species, we have to reject—or at least inspect—the framing of the inevitability of AI. Unavoidable for the most part, yes. But whose agenda is being slipstreamed in the blind acceptance of this future, as dictated by the players of big tech?
Because in my mind, the technology itself is not inevitable. AI is only as inevitable as the story and the people it can convince of its purpose.
And I would say that story—more and more compelling as more influential people are compelled by economic or personal pressure—is being "written in" to the paperwork of the inevitable.
A technology met a market where specific people were compelled to try to change the world. And the world was configured for a system uniquely susceptible to that compulsion.
The question is: are you going to let them write you into it too?
Thanks for reading, see you in the next one.
Ia manuia.
— Ron