demoss
Explorer
Technology is always a force multiplier.
This is just more of the same, but admittedly things are coming to a head; the multipliers are getting pretty big. We need to build societies and systems that are robust in face of such force multipliers.
The only reason anyone does anything, is because they have an incentive to do it. To get more good outcomes and less bad outcomes we can:
As a global society we're utter crap at this.
Particularly, we're almost never thinking about effects of perverse incentives and perverse punishments: our political and economic systems are chock full of them, but they're taken for granted.
Example: scientific publishing is going in disarray as AI generated completely fictious papers have started appearing even in prestigious journals (turns out LLMs are scarily good at creating plausible sounding), making the already bad combination of publication bias and replication crisis across most disciplines even worse. Publish-or-perish model of research and profit seeking model of scientific publishing have created a perfect cycle of perverse incentives where (1) only successful research gets funded (2) success of research is measured by number of publications (3) publishers are motivated profit, more publications means more profit (4) challenges to the models of funding and profit are penalized.
(I'm hoping this is still safely in the realm of philosophy instead of polics...)
This is just more of the same, but admittedly things are coming to a head; the multipliers are getting pretty big. We need to build societies and systems that are robust in face of such force multipliers.
The only reason anyone does anything, is because they have an incentive to do it. To get more good outcomes and less bad outcomes we can:
- Create rewards for doing good. (incentives)
- Remove rewards for doing bad. (perverse incentives)
- Create penalties for doing bad. (punishments)
- Remove penalties for doing good. (perverse punishments)
As a global society we're utter crap at this.
Particularly, we're almost never thinking about effects of perverse incentives and perverse punishments: our political and economic systems are chock full of them, but they're taken for granted.
Example: scientific publishing is going in disarray as AI generated completely fictious papers have started appearing even in prestigious journals (turns out LLMs are scarily good at creating plausible sounding), making the already bad combination of publication bias and replication crisis across most disciplines even worse. Publish-or-perish model of research and profit seeking model of scientific publishing have created a perfect cycle of perverse incentives where (1) only successful research gets funded (2) success of research is measured by number of publications (3) publishers are motivated profit, more publications means more profit (4) challenges to the models of funding and profit are penalized.
(I'm hoping this is still safely in the realm of philosophy instead of polics...)