3 ms·
It won't. It will get better. In a couple of years you will be able to take a picture of a rash in your arm and get exact diagnosis with treatment instructions
by Entangled 10y ago
It won't. It will get better.
In a couple of years you will be able to take a picture of a rash in your arm and get exact diagnosis with treatment instructions. You will be able to take a pic of a flower, a leaf or any tree and get accurate info about its species, plagues and best techniques for growing them. You will be able to take a pic of any insect, spider, snake, animal, or anything that moves, any mineral, element or anything at all and get accurate info about that.
AI is not only about robots thinking, it is about collecting information and making it available on demand. Agriculture and health will be the first beneficiaries, finances in a close second.
Data mining is where the first stage of AI is, and Google is moving ahead of everybody else with their search engine, maps, translation, and all the information collecting tools. Once you have enough data, knowledge is just a couple of programs away.
- IshKebab 10y agoThat is a valid use of deep learning. I think the article is talking about people thinking that because deep learning is very impressive for some things, they should use it for everything - even problems where a much simpler solution works fine, or problems where deep learning doesn't apply. I work for a consumer product company, and there are often people talking about using 'big data' and 'machine learning'. They're just following the hype; they don't really know what machine learning is and I've only heard two or three potential applications of it mentioned that make any kind of sense.