The value of AI to business, individuals, and society in general is infinite productivity gains.
Yes, I mean it. Not hyperbole. Not a number I pulled out of the air. It is demonstrably a fact, and one I can attest to from personal experience.
Here is how I arrived at it.
Before AI
- Music produced: 0
- Python utilities written: 0
- Old documents indexed and catalogued: 0, was never going to get around to it
- Archived and indexed document library: 0, too much grinding, would never do it
- Non-critical but time-consuming server problems fixed: 0
After AI
- Music produced: 4 albums, 42 tracks
- Python utilities written: 25
- Old documents indexed and catalogued: 7,234
- Archived and indexed document library: done. $16 of tokens, well spent
- Non-critical but time-consuming server problems fixed: dozens. Another good use of $8 of compute
The maths is straightforward. Zero is the floor. Anything above zero is, by definition, infinite gain.
Now, I can hear the objection already. “But Steve, That isn’t real productivity gain, that is just work you weren’t doing.” Fair point, on the surface. First off, no: the coding and the music production was work I could not do at all without AI assistance. I am not a musician, I am not a coder. But now I can produce both.
Now look at it again. These are not idle wishes. These are things I actually wanted done. Things that had business value, creative value, or personal value to me. Things I had thought about, planned, even started, and then put aside because the cost in time and effort was too high relative to the available hours in the day.
That backlog is not unusual. Most people I know have one. A pile of useful work that never quite makes it to the top of the list, because the top of the list is occupied by the urgent rather than the important. AI does not just help with the work you are already doing. It clears the work you were never going to get to.
That is the part most of the commentary misses.
The conversation about AI productivity tends to focus on the office worker writing emails 20% faster, or the developer shipping code 30% faster. Those gains are real, but they are incremental. They are the visible part. The much larger gain, in my experience, is the work that simply never would have happened, now happening.
Four albums. I am not a musician. I have always wanted to make music. Without AI, that was never going to be more than a vague intention.
Twenty-five Python utilities. I can read code well enough, but I am not a developer. Each of those utilities saves me time or removes friction from something I do regularly. Compounding gain, paid for once.
Seven thousand documents indexed. That is twenty years of business and personal correspondence, blog exports, technical notes, and project records, now searchable and useful instead of sitting in folders I would never open.
The cost of all this, in actual dollars to the AI providers, is trivial. A few hundred dollars across a year of fairly serious use. The cost in my time is real, but it is the kind of time I want to spend, not the kind I resent.
So to all the flint knappers, farriers, lamp lighters, switchboard operators, and the rest of the professions that bemoaned the technology that replaced them: sorry dudes. There is another job for you somewhere, being more productive than you ever imagined.
Look, I understand the worry. Some of it is justified. The transition will be ugly in places, as transitions always are. But the productivity gain at the individual level is not theoretical. It is not coming. It is here. The bottle neck will be the ability of organizations to soak up and make use of the extra productivity. It is here. I am living in it.
The only question is whether you are using the time it gives you, or watching someone else use theirs.
