http://www.industryandbusiness.ca/insight/demystifying-the-value-of-ai

Yes, it is but it is my believe that the knowledge of man the thinker will always be superior being the creator

https://futurism.com/artificial-intelligence-hype

The answer can be yes and no, depending on what you mean by "smart enough". Yes, because it is assisting humans in decision making tremendously. No, because of so many reasons, e.g., it is yet a long way from incorporating feelings, it is still dependent on humans (aka domain experts) to its being "smart enough" and many more.

Think about it this way. How human experts became system experts? the answer is by learning... even humans themselves can be more expert in some fields than others, based on their educational level and experience. So no expert is an expert without learning from another expert! Same applies to AI or Machine learning techniques, the more they learn (using expert knowledge or any other sources of knowledge) the best they become in fulfilling the end goal they built to do. However, there are many online/active machine learning approaches that doesn't require any prior knowledge or training! While humans always do require training.

I believe that AI techniques have rapidly become way more sophisticated and advanced than before. And to answer your question is; yes! there are many approaches to machine learning. Most specifically, online/active unsupervised methods which can learn entirely without any labels, pattern or learning context... that is so similar to regular human expert patterns of learning. And these techniques can even beat human experts in their precision and accuracy.

I think the real question should not be about the accuracy/abilities of AI comparing humans. Instead, I believe the real question should be: Could we create machines with enough computational power to handle all these complex and high-dimensional operations. To perform these tasks, store them, and response to them as fast as the human brain does?

That's why deep learning is more famous than other techniques.... The fight now between all these methods is about computational power, storage power and speed, not only the ability or accuracy anymore.

In his Ted Talk, Garry Casparov puts Machine vs Human Intelligence pretty precisely. https://www.ted.com/talks/garry_kasparov_don_t_fear_intelligent_machines_work_with_them?language=en

thanks Sisay Chala , it was an interesting TED!

As always machine's triumph is a human triumph! Something we tend to forget when humans are surpassed by our own creations.

We must face our fears if we want to get the most out of technology -- and we must conquer those fears if we want to get the best out of humanity, says Garry Kasparov. One of the greatest chess players in history, Kasparov lost a memorable match to IBM supercomputer Deep Blue in 1997. Now he shares his vision for a future where intelligent machines help us turn our grandest dreams into reality.

No ,there is more should be done .

What I am thinking in the perspective of application domain, I mean it depends on where we can apply AI. For example, to check patient's normal daily check-up AI may be performing well with current ML techniques, while at the same time we are not sure about AI applications in the field of Autonomous Vehicles (AVs) especially, after fatal accidents of Tesla and Uber. Altogether, one has to defined well about the effective application domain of AI, and accordingly we should take the next step for implementing the technology based on AI.