3 ms·

"The latest estimates, using official figures, suggest that real output per employee in the median rich country is not growing at all." If I can use an AI assi

"The latest estimates, using official figures, suggest that real output per employee in the median rich country is not growing at all." If I can use an AI assistant to complete a job in 3 hours that used to take 9 hours, that's a win for me, because now I have 6 hours to spend on a personal open-source project, some family business, recreational (and healthy) exercise, or some community-beneficial volunteer activities. While this might not show up on the corporate balance sheet, it's going to make the overall society a lot healthier. The existential risk of AI to the current socioeconomic order might have more to do with AI systems making better economic decisions than the overcompensated executive-shareholder sector does, and at far lower cost to the corporation or institution that adopts it. This makes logical sense, since the skills involved in humans climbing to positions of power in organizations (manipulation of other people, hyping up one's capabilities, clinging desperately to power, etc.) don't translate into efficient well-run operations.

purple-leafy 2y ago
Except that’s not how it ever works. Any productivity gains will be consumed by your employer, and the expectations of your position will increase with the productivity gains. Ie: if your role is now 75% more productive on average, the new expectation of that role is a 75% increase in outputs. So it actually will get much worse for you, because you’ll be doing much, much more for effectively the same pay :) There used to be a single role that did the data layer, a single role that did QA/Testing, a single role for UI/UX, a single role for client side, a single role for server side, and a devops team. Then when the world became good at this stuff, the expectation shifted to, in many cases, a single role juggling all of this work
ghaff 2y ago
It sort of depends on what you assume become table-stakes tools. Spreadsheets reduced some work, created more work (run a bunch of alternative models), and yes probably eliminated some work--for the people fluent in them. It's probably hard to argue that spreadsheets (which have become the most inarguably successful low-code tools) have markedly decreased the amount of work their users need to do in companies.

purple-leafy 2y ago

Except that’s not how it ever works. Any productivity gains will be consumed by your employer, and the expectations of your position will increase with the productivity gains. Ie: if your role is now 75% more productive on average, the new expectation of that role is a 75% increase in outputs. So it actually will get much worse for you, because you’ll be doing much, much more for effectively the same pay :) There used to be a single role that did the data layer, a single role that did QA/Testing, a single role for UI/UX, a single role for client side, a single role for server side, and a devops team. Then when the world became good at this stuff, the expectation shifted to, in many cases, a single role juggling all of this work

ghaff 2y ago

It sort of depends on what you assume become table-stakes tools. Spreadsheets reduced some work, created more work (run a bunch of alternative models), and yes probably eliminated some work--for the people fluent in them. It's probably hard to argue that spreadsheets (which have become the most inarguably successful low-code tools) have markedly decreased the amount of work their users need to do in companies.