Not sure why everyone is saying no, the answer is obviously yes. Having infinitely patient and accurate brains that can spin up in a dime will be the greatest economic boon we have ever seen.
Absolutely nothing about our current generation of AI is accurate at all.
67% even 87% on synthetic benchmarks does not intelligence make.
It’s all statistics based, it’s not infinitely accurate, nor do we have any reason to think any AI system would exhibit anything resembling patience. They don’t exist outside of inference time, let alone have a sense of the passage of time.
Sure today it is, what about in a year or two years? And yes AI models are way more patient than humans for the exact reason you mention, the logistics of that don't matter, the result does
GDP growth doesn't necessarily mean any one person will be better off. But regardless, if you, say, build something with the help of AI that you couldn't do before, it could make you money, depending on how much you sell.
The ease with which he can use AI is the same as for everyone else. Therefore he gains no relative benefit and can't "make money from AI".
However, information-retrieval and "thinking assistance" helps everyone at the same time. Asking for summarization and clarification of any (legal or technical) document, finding the right references, and so on. It's like everyone gets instant access to a research department.
This means fewer people are needed for "clerical jobs". Effectively, it diminishes the economic value of memorisation and mechanical information work. But that's probably good a thing.
> The ease with which he can use AI is the same as for everyone else. Therefore he gains no relative benefit and can't "make money from AI".
Not really, people are not equal in output. Some people can use their tools much better than others, like juniors versus master craftspeople. Even in startups we see some succeed while many fail, and while there may be macroeconomic conditions (or otherwise those outside of one's control) for such failure, some still rest on the founders themselves and their decisions.
depends. The issue is that LLMs effectively “centralize” the functionality into a small set of models, which means if the costs of service drop too much compared to increase in demand, then GDP may perversely decrease. It all depends on whether AI actually increases net demand
Absolutely nothing about our current generation of AI is accurate at all.
67% even 87% on synthetic benchmarks does not intelligence make.
It’s all statistics based, it’s not infinitely accurate, nor do we have any reason to think any AI system would exhibit anything resembling patience. They don’t exist outside of inference time, let alone have a sense of the passage of time.