Everywhere I look, people are talking about AI. Some are very optimistic. Some are very pessimistic. Most of them are loud.
The other day I had a comforting thought. Maybe it is good that I am sitting in the driver’s seat. To use the fucker’s power, you need domain knowledge. You need to know what to ask, and you need to know when the answer is wrong. If I keep that, it will keep me on top.
The day after that, I was stuck on a problem with a sequencing library. Properly lost. I gave it to AI, and it ran a bunch of analysis that would have taken me a few hours, if I had even known I should do it. It took 10 minutes. At the end, it handed me a list of references.
So much for staying on top. That got me thinking about Sartre:
“Man is nothing else but that which he makes of himself… man is nothing else but what he purposes, he exists only in so far as he realizes himself, he is therefore nothing else but the sum of his actions, nothing else but what his life is.”
AI will become what that “man” needs it to be: a proper scientist, or whatever else we need, with deep domain knowledge, that does not sleep, thinks very fast, and has retrieval power that is almost unimaginable for a human. It is getting there, and given enough time, it will be there. Simply put, my tokens per second are way lower than a regular AI’s.
I was anxious for a few days. I kept thinking about how to make myself indispensable. Learn more. Learn faster. Find the corner of the work the machine can’t reach yet, and stand in it until it reaches that too.
It wasn’t a good feeling. It felt like running on a treadmill and someone else holds the speed button.
And then I remembered: who cares?
AI, no AI. Job security, insecurity. Que sera, sera.
I am part of the human colony. I will be there, at the edge, watching from the cliff, seeing what happens. Or I will fall, and then I will know how that feels too.
Good or bad, I am here to see how it goes.