4 ms·
Maybe she has inside info that it's not "around the corner". Making bigger and bigger models does not make AGI, not to mention exponential increase in power req
by romanovcode 2y ago
Maybe she has inside info that it's not "around the corner". Making bigger and bigger models does not make AGI, not to mention exponential increase in power requirements for these models which would be basically unfeasible for mass market.
Maybe, just maybe, we reached diminishing returns with AI, for now at least.
- steinvakt 2y agoPeople have been saying that we reached the limits of AI/LLMs since GPT4. Using o1-preview (which is barely a few weeks old) for coding, which is definitely an improvement, suggests there's still solid improvements going on, don't you think?
- samatman 2y agoContinued improvement is returns, making it inherently compatible with a diminishing returns scenario. Which I also suspect we're in now: there's no comparing the jump between GPT3.5 and GPT4 with GPT4 and any of the subsequent releases. Whether or not we're leveling out, only time will tell. That's definitely what it looks like, but it might just be a plateau.
- xabadut 2y ago+ there are many untapped sources of data that contain information about our physical world, such as video the curse of dimensionality though...