There is a huge ambiguity in the goals of artificial intelligence (AI) and machine learning used today by large companies that desire to create an artificial intelligence that could perhaps match or even exceed the capabilities of the human mind.
As they develop, artificial intelligence or AI and machine learning or machine learning will be able to take on even more complex tasks, but it could take half a century or more before any AI reaches human-level intelligence at satisfactory levels. We have only seen the sequel in disaster movies, but what is certain is that super-intelligence seems to excite some, and terrify others. In the meantime, the new technology has continued to fuel science fiction for decades (see Skynet).
Currently, today's AI (artificial intelligence) helps companies improve customer service or refine their decision-making by identifying trends in data that would otherwise be invisible. AI helps automate routine tasks, or even create entirely new services.
From the above it follows that any AI technology should consider several issues:
Artificial intelligence is a rapidly developing and exciting technology, but it is not the answer to every problem. In particular, watch out for the “AI” prefix in future product names because it does not necessarily mean that it is better than another one that does not contain the prefix.
Furthermore, the lack of skilled staff who can make the most of the technologies, along with massively inflated expectations, could create a loss of trust.
But perhaps the most dangerous thing is to treat any AI as magic. An algorithm is only as good as the data or rules that human programmers feed it. The nature of algorithms that can learn and evolve without programmers will be slow to come, and again, that doesn't mean their answers will be accepted without question.
Instead, we need to find ways to make sure that AI decision-making is constantly challenged. Challenge is the beginning of all science and all scientific discovery. So some researchers are looking at using a lot of factors, such as accountability, explanation, accuracy, controllability, and fairness. As you can see, because the concepts themselves are abstract and not governed by specific laws, it takes a lot of work.
It is also important to consider the impact of artificial intelligence in its broader sense:
These technologies will have the potential to significantly alter some jobs, create some and destroy others. Developers of this technology and its users will need to consider and acknowledge the potential positive and negative consequences. AI-powered autonomous vehicles, for example, may reduce pollution and make travel more enjoyable – but they will also throw many drivers out of the job market. So we need to understand more broadly and have more conversations about the coming changes.
As it turns out, artificial intelligence and machine learning are not something we need to worry about. Our concern should focus on human intelligence and our own ability to learn and predict the future.
