2 ms·
I don't see any reason to believe that just because the idea failed in implementation once that it will forever. In all likelihood, if we try at the problem lon
by tensor 10y ago
I don't see any reason to believe that just because the idea failed in implementation once that it will forever. In all likelihood, if we try at the problem long enough someone will eventually figure out algorithms that work fairly well.
Perhaps "AI design" competitions or benchmarks would help this line of work. But people trying multiple times can only be a good thing, even if the early attempts are not great.
- Nadya 10y ago>Due to it being A.I-based the only thing I can really say is "It's been tried before. Best of luck on doing it better." I'm sincere in saying that. But unless you somehow draw up some marketing hype - it becomes increasingly unlikely to receive funding for an idea that's failed in the past unless you have some tangible evidence about how you're avoiding running into the same problems. If (/when) they have an actual product we'll see if they managed to do it better. But there have already been enough failed attempts (TheGrid wasn't the first or only one since) to not be hyped because "AI codes my site". So with nothing else to go off of other than "This is attempt #x by some new people (or person)" I can't say I'm excited or have high expectations. I'd be lying.