4 ms·
It is hard for me to not think that given the current approach it will always be almost there, but not quite good enough to be more than a nice assistant or too
by etempleton 2y ago
It is hard for me to not think that given the current approach it will always be almost there, but not quite good enough to be more than a nice assistant or tool that has to be meticulously built and maintained by a company. Take, for example, self driving vehicles, we have been 5 years out for 25 years. That is not to say there has not been significant progress, but to go from a really powerful driver assist to full self driving has still not really come to full fruition and that is kind of a best case scenario for AI/ machine learning.
That is just one very narrow task that basically anyone can do regardless of intelligence or talent. It takes less than a year to train a distracted 16 year old to do it, but three decades to train an AI and even then you probably need to hand tune it for specific locations because it won’t know what do in unusual road layouts.
I think we will get there with driverless vehicles, but only because it is being tailored by humans to handle all of the edge cases. Self driving cars, if truly self driving with no user intervention, will of course be worth it in the end. The trillions of dollars spent will eventually add countless dollars in added efficiency and unlimited revenue for those who get there first, but how many applications can we really say that is true for? That the decades of development and training it takes to remove humans will be worth the initial investment?
I guess my contention is that the current development path seems unlikely to ever achieve AGI and so instead you will have to do heavy customization to get anything that is much more useful than what we have now.
- andrewchambers 2y agoI feel like your point would be better if there weren't already self driving cars around like waymo.
- ileonichwiesz 2y agoExcept they’re not fully self-driving, they’re all at least partially driven by an underpaid contractor in a remote warehouse.
- etempleton 2y agoThat is kind of my point. They exist only in a few select locations. If they were truly self driving Google would be rapidly rolling them out around the World, but they are not.
- Closi 2y agoWaymo's first real driverless ride on a public road was less than 10 years ago. Now they are doing 100k rides a week. I think your original post just overestimates the 'normal' pace of change - Taking 10 years to go from the first ride to 100k rides a week is no time at all in the scheme of things. Airplanes are an example of an insanely fast technological adoption, and they still took 30 years to go from the wright brothers to the first commercial jet. I get the feeling that if the wright brothers invented the plane in 2024 though, people would be saying 10 years after "Planes are all hype! If planes were truly transformative they would be rolled out everywhere and would be carrying passengers by now"
- gcanyon 2y ago> 30 years 45 years: December 1903 to July 1949 https://en.wikipedia.org/wiki/Wright_Flyer https://en.wikipedia.org/wiki/Wright_Flyer https://www.history.com/this-day-in-history/first-jet-makes-test-flight https://www.history.com/this-day-in-history/first-jet-makes-...
- etempleton 2y agoI am not debating the transformative nature of self-driving cars or if it will become common place. I think it is huge. I guess my point is that is just one application of machine learning / AI. To automate such a complex task with our current approach to AI, that humans do easily, one will need to spend trillions of dollars and therefore the benefit, such as with self-driving cars, needs to be significant. I suspect this will be the case with any AI workload. Heavy customization to get to a place where it is good enough. So the financial benefit needs to be equally huge and someone needs to feel confident they can extract that value 20 years in the future.