4 ms·

What happened to self-driving for example, though? The growth of the the technology hasn't been exponential, it seems to grow under a diminishing marginal retu

What happened to self-driving for example, though? The growth of the the technology hasn't been exponential, it seems to grow under a diminishing marginal returns law, that is, the first 90% was "easy" to solve, but every bit of the remaining 10% seems increasingly and excruciatingly harder. If you're observing local behavior in the first 90% of the curve it might look like it could explode any time but unbeknownst to you, the behavior towards infinite is a actually a constant ceiling. I'm not categorically denying it's exponential, I'm just saying it's very difficult to make assumptions based on observed local behavior.

vidarh 4y ago
Self driving is limited by the only real testing requiring real hours of driving with real cars, where getting the last 10% of the way requires getting real-life testing in increasingly rare conditions.
- arcturus17 4y ago
  But can you ascertain that similar dynamics won't also come into play for software? Yes, software is easier to deploy and test at scale - there are barely any physical constraints compared to driving. But on the other hand it's also a more complex domain than driving. Taking Advent of Code as an example, GPT seems to be making mince meat of some problems, but it struggles with others. Is it not conceivable that in software there will be a "last 10%" where AI will trip up, either because the inputs (prompts/requirements) are hard to express or parse, or the output (code) is hard to generate?
  - vidarh 4y ago
    "Similar dynamics" requires situations where training the AI becomes bottlenecked on training opportunities. There will certainly be areas where such bottlenecks exist, and so we can assume that there will be types of problems we can't expect exponential improvement to continue once it hits those bottlenecks. But the point is that self-driving is a poor example to use to judge what we can expect to see in fields that are less constrained by turnaround times and safety. Not least because "anyone" - though certainly constrained by costs - can try to do novel things with large language models. Very few can try to do novel things with self-driving because they have nothing realistic to test it on at scale.
- phkahler 4y ago
  No. Full self driving requires development of full AGI. General intelligence is needed to navigate construction zones with a flag man directing traffic someplace it normally doesn't go. As but one example. Stopping to ask said flagman a question for clarification is also preferable to just guessing the answer.
  - kadoban 4y ago
    There are levels of self-driving far below that that capture almost all of the useful improvement.
  - vidarh 4y ago
    While I don't agree with what you write here, none of it contradicts what I wrote above.
cma 4y ago
Cruise has 100 operating simultaneously in SF with no safety driver: https://www.linkedin.com/posts/olivercameron_speaking-of-ai-breakthroughs-this-is-what-activity-7004526630786269185-SatQ https://www.linkedin.com/posts/olivercameron_speaking-of-ai-...
- arcturus17 4y ago
  100 cars driving around the tech capital of the world is not even close to the scale many people imagined 5-10 years ago though.
  - fsflover 4y ago
    Do you expect that it will stop here? People usually overestimate the short-term changes and underestimate the long-term ones.
    - arcturus17 4y ago
      No, but the last five years of progress have not been exponential, which is the whole point of discussion with GP. A major breakthrough could see it 99.9% solved and adopted at scale, but until then it could be a slow grind for every 0.1% of progress.
      - cma 4y ago
        Driving frequently in a large areas of a busy city without a safety driver is insane progress. Just lots of hype-men like Musk (and earlier Jensen) made wild promises.
- ac29 4y ago
  It should be noted this service is only available 10PM-530AM, and only in the less traffic-y areas of SF: no downtown/market, mission, etc. https://getcruise.com/rides/ https://getcruise.com/rides/
  - 4y ago

vidarh 4y ago

Self driving is limited by the only real testing requiring real hours of driving with real cars, where getting the last 10% of the way requires getting real-life testing in increasingly rare conditions.

arcturus17 4y ago
But can you ascertain that similar dynamics won't also come into play for software? Yes, software is easier to deploy and test at scale - there are barely any physical constraints compared to driving. But on the other hand it's also a more complex domain than driving. Taking Advent of Code as an example, GPT seems to be making mince meat of some problems, but it struggles with others. Is it not conceivable that in software there will be a "last 10%" where AI will trip up, either because the inputs (prompts/requirements) are hard to express or parse, or the output (code) is hard to generate?
- vidarh 4y ago
  "Similar dynamics" requires situations where training the AI becomes bottlenecked on training opportunities. There will certainly be areas where such bottlenecks exist, and so we can assume that there will be types of problems we can't expect exponential improvement to continue once it hits those bottlenecks. But the point is that self-driving is a poor example to use to judge what we can expect to see in fields that are less constrained by turnaround times and safety. Not least because "anyone" - though certainly constrained by costs - can try to do novel things with large language models. Very few can try to do novel things with self-driving because they have nothing realistic to test it on at scale.
phkahler 4y ago
No. Full self driving requires development of full AGI. General intelligence is needed to navigate construction zones with a flag man directing traffic someplace it normally doesn't go. As but one example. Stopping to ask said flagman a question for clarification is also preferable to just guessing the answer.
- kadoban 4y ago
  There are levels of self-driving far below that that capture almost all of the useful improvement.
- vidarh 4y ago
  While I don't agree with what you write here, none of it contradicts what I wrote above.

cma 4y ago

Cruise has 100 operating simultaneously in SF with no safety driver: https://www.linkedin.com/posts/olivercameron_speaking-of-ai-breakthroughs-this-is-what-activity-7004526630786269185-SatQ https://www.linkedin.com/posts/olivercameron_speaking-of-ai-...

arcturus17 4y ago
100 cars driving around the tech capital of the world is not even close to the scale many people imagined 5-10 years ago though.
- fsflover 4y ago
  Do you expect that it will stop here? People usually overestimate the short-term changes and underestimate the long-term ones.
  - arcturus17 4y ago
    No, but the last five years of progress have not been exponential, which is the whole point of discussion with GP. A major breakthrough could see it 99.9% solved and adopted at scale, but until then it could be a slow grind for every 0.1% of progress.
    - cma 4y ago
      Driving frequently in a large areas of a busy city without a safety driver is insane progress. Just lots of hype-men like Musk (and earlier Jensen) made wild promises.
ac29 4y ago
It should be noted this service is only available 10PM-530AM, and only in the less traffic-y areas of SF: no downtown/market, mission, etc. https://getcruise.com/rides/ https://getcruise.com/rides/
- 4y ago