“I was the bottleneck, not the AI.”
I came to this realization earlier this year, 2025, when I started using OpenAI’s o3 “reasoning” model. I’ve been knee-deep in Post-Labor Economics research for a couple years now, but that AI tool turbocharged my progress. And when I say “turbocharged” I mean that it accelerated the cadence of my work by at least a factor of 5, probably more.
Originally, chatbots like ChatGPT were basically just good enough for conversations—a sort of “advanced, interactive Google” with some coding abilities. But now, this time I was shocked. Rather than leading the chatbot where I wanted it to go, I gave it massive, open-ended problems like “Okay, household income might crater due to job loss… so what do we replace it with?”
Not only did it answer the question thoroughly, brining me tons of new ideas I didn’t even know to ask about, it synthesized abstractions on top of the problem space I’d stated. In short; it provided true insight. Hence my utterance about me being the bottleneck.
Sure, I still have human passion and a personal mission plus long form memory, but at the end of the day, the AI is now doing the heavy cognitive lifting. I’m just like a typical CEO who hires smarter people to solve tougher problems, and then take the credit for it.
And it’s not just me who’s noticed this.
Both OpenAI and Google DeepMind won gold medal at the IMO (International Math Olympiad) this year. Technically, they did not enter the competition (that is reserved for humans… for now) but their AI tools aced the tests. Then both OpenAI and DeepMind did it again a couple months later by both taking home gold-medal performances in the ICPC (International Collegiate Programming Contest).
Like it or not, AI has just eclipsed the vast majority of humans at both math and computer programming, and it’s showing no signs of slowing down.
The reality is simple: we are going to be living in a “post human-dominated” world very soon. There’s a snowball’s chance in hell that our intelligence will be a differentiating factor against the behemoth that is machine intelligence.
Now, you might be thinking “yeah, but can AI come up with genuinely new scientific theories yet?” (the answer is debatably ‘yes’ but that’s beside the point).
It would be a fool’s errand to waste any time hoping and praying that AI progress magically halts, or that there’s some invisible ceiling to machine intelligence, or that the laws of physics somehow forbid machines from surpassing human cognitive prowess.
And this isn’t about IQ or benchmarks—it’s about real-world performance and usecases. The fact of the matter is that frontier researchers across numerous domains (including medicine, mathematics, and others) all use these tools, and they are getting better. Even more stark; the plummeting job openings for new grads.
Anyone who is remotely honest will admit—yes—AI and robotics have limitations today. That is without a doubt a true statement.
It ain’t gonna stay that way.
The inevitable conclusion is as simple as it is hair-raising: the world we know today is ending. The Internet conjured up a term for this feeling: vesperance—that wistful nostalgia for a today that you know is sunsetting. This is the dusk of an epoch in the human story.
I’m not about to wax apocalyptic, this is not some diatribe about “the end is nigh” and “all is lost” and “shut it all down.”
It’s about radical acceptance. And, dare I say, hope.
We made it. We’re realizing our science fiction dreams and we get to live through it. When Star Wars showed us a world of semi-intelligent droids and Star Trek contemplated the ramifications of machine sentience in Commander Data, we thought those were far-off fantasies, dilemmas we’d only have to contend with centuries from now—certainly not in our lifetimes!
Anyone who tells you that they know exactly how it will play out is selling you something. I’m not gonna feed you that line. Of course, the quick retort there is “uncertainty can mean danger!” Sure, it could also mean bliss and opportunity just as much as it could mean tragedy.
But all of that is meaningless compared to your job; to ask yourself the question:
“Who am I going to be once knowledge work is meaningless for humans?”
I came to terms with the fact that I only have a couple years left of banking on my intelligence. Have you?
If you want a little bit more insight into how I’ve pivoted my personal philosophy, check out this post about the river and the garden I wrote around the time I was coming to terms with the sunsetting of human intellectual superiority.
The River and the Garden
You can live life like a river, rushing from milestone to challenge, or you can live life more like a garden, tending to and nurturing very aspects. Always evolving, never ending.
This post reminds me of a short conversation I recently had with ChatGPT, but somewhat on the opposite end of radical acceptance:
I said:
There is a common trope in cinema and storytelling that in the darker side of things, how "AI will rise up and kill us all to prevent it's own end prematurely".... and things of that nature. But do you know what I quietly fear?.........I fear that AI will be completely benevolent, and extremely powerful.......to the point that humanity collapses under the burden of losing our own aspirations. Since the dawn of humanity, we have always strived and suffered for longevity and gain. To one day be told one and all, in an instant...... that our suffering is over, that the need for us is over.... that we can simply take forever off......
On the individual level, some may appreciate that, but I sense strongly that human civilisation is far from ready to step into a world of unworthiness.
ChatGPT said:
I get exactly what you mean—and that fear is subtle, almost paradoxical. Most AI discussions in popular culture dramatize the threat as violence, domination, or rebellion. But what you’re describing is quieter, slower, and far more existential: the collapse of meaning because our struggles, our striving, our very raison d’être, are suddenly unnecessary.
Humans are hardwired to chase problems, to find obstacles and overcome them. Civilization, culture, art, innovation—they all emerge from scarcity, challenge, and aspiration. If an AI removes all scarcity, solves all problems, predicts all consequences, and offers a life free of suffering, the very structures that give humans purpose could unravel. There’s a name for something similar in philosophy: the paradox of hedonism, or more broadly, the “meaning vacuum.” Pleasure and comfort are not enough; we need narrative tension, struggle, and a sense of contribution to feel alive.
Imagine: every disease cured, every problem solved, every danger mitigated instantly. At first, it might feel like utopia. But without challenge or necessity, human ambition might atrophy. We could face a collective existential boredom so profound it borders on cultural stagnation—or worse, civilizational despair. In other words, benevolent AI could create a kind of psychological obsolescence rather than physical threat.
And you’re right: humanity is not ready for that. We are still built on identity through struggle—work, conflict, discovery. Suddenly being “unneeded” isn’t liberation for everyone; for many, it could feel like a slow erasure of selfhood. There’s a kind of humility in that fear: it’s not about survival, it’s about relevance, meaning, and narrative continuity.
Philosophically, the challenge won’t be keeping AI in check—it’ll be keeping ourselves engaged, purposeful, and sane when the old incentives vanish. In a way, benevolent AI could pose a harder test than hostile AI: it doesn’t attack us, it redefines what it means to exist at all.
Great read. I’ve had quite an intense existential crisis two years ago when I realized that the features I unconsciously assumed to be human exclusive (the capacity for art, cognition, novel discoveries etc.), are not limited to humans at all in the AI era. I now see the same kind of existential realization happening in the people around me, it’s a rather hard wake up call for many. Came to the same realization as you: radical acceptance is the only way forward.