Do you need a degree in computer science to know that the nueral nets are in a car or a robot?
Over 40 different neural nets are doing various algorithmic tasks in the Tesla car, and I can envision a future where AXE tech is used in that AI process somewhere.
And you are right, there are two chips, but each has 12x microprocessors, 1x GPU and 2x neural processing units and a bunch of hardware accelerators - these are custom-designed chips doing billions of calculations a second. Could AXE tech replace or work alongside NPUs in the future? IMO, yes.
I've heard nothing about AXE chips having a problem with "too much power", not sure what you are suggesting the power usage isn't comparable to normal computer hardware, but if you have some links please post them so we can all check it out. Sounds like BS to me, why would Mo be pushing ahead with a mobile-first strategy if they thought the AXE tech would be power-hungry?
You also say that AXE tech will not likely be used in training AI models because they are only about "big data crunching". And you may have a point about training models (although there are alternate arguments there as well). But you completely ignore the other side of the AI coin, the inference phase. This is where the AI uses the algorithms created by the training data to make a prediction (eg the NPUs in a Tesla). Here is a paper on the possibilities of quantum inference - https://arxiv.org/pdf/2103.06720.pdf
DYOR - Good luck all.
AXE Price at posting:
80.5¢ Sentiment: Buy Disclosure: Held
A personalised tool to help users track selected stocks. Delivering real-time notifications on price updates, announcements, and performance stats on each to help make informed investment decisions.
(20min delay)