In my last post , I demonstrated by means of the Mathematica® software that artificial intelligence does not always meet satisfying or even correct statements. I have done this to help you get started with this post.
The changing, digitized future needs different, sometimes new, skills from people in dealing with new technologies.
Artificial intelligence with its subsets of machine learning and deep learning is based in many applications on training the neural networks or algorithms behind it. First of all, they must be feeded with given knowledge in the form of data, whereupon a model of reality is formed in the computer. New data, unknown to Artificial Intelligence so far, are processed with the models and are then available for further processing or use.
Simply put, the better the training and the better and clearer the new data, the better the results.
Here is an example from my previous calculations with Mathematica® . We found that determining the geographical position in particular led to problems. No wonder, because the amount of information about the location in the background of the picture was very small. It is different in the second example. A shot of me with a section of the Brandenburg Gate and belonging Quadriga in the background. For the calculation the identical algorithms were used.
excellent result: Berlin, Brandenburg Gate
Such problems can occur in many ways in future. It is up to the human to decide if the result corresponds to reality. Of course this should not happen all the time.
However, this raises the question of what skills people will need to "survive" in the digital future. The "World Economic Forum" in cooperation with "Guthrie-Jensen Consultants" has created an excellent graphic with all the necessary information.
by courtesy of Guthrie-Jensen Consultants
Here are some skills that experts say should be prioritized:
1. Complex Problem Solving
It’s true that AI can solve problems that humans cannot – but it also goes the other way. When problem solving needs to span multiple industries or when problems are not fully defined, humans can work backwards to figure out a solution.
2. Critical Thinking
Machines are getting better at aspects of critical thinking, but humans are still able to to connect, interpret and imagine concepts in a world full of ambiguity and nuance. A lawyer can pinpoint the exact positioning to make a case for a client, or a marketer can figure out an overarching message that can resonate with consumers.
However, critical thinking is also of great importance with regard to the examples shown here, for those who expect that the results can always be considered unconditionally correct can sustainably draw false conclusions or even make wrong decisions .
Creativity requires a degree of intuitive randomness that can not yet be imitated by AI. Why did the architect design the building a certain way, and why did the musician improvise by playing a chord out of key? It’s hard to explain why to a computer – it just feels right.
Now, looking back at my first Mathematica® example  and envisioning that this is a result of your work environment, which of the skills listed here do you need to understand the results and communicate the results internally or externally? Do you need more skills?
Do you have questions or comments about this post? Then please send me an email. I am looking forward to hearing from you.
 Jürgen Kanz, "Can we trust the results of Artificial Intelligence without a critical view? ", https://www.juergen-kanz.de/en/blog/index.php?can-we-trust-the-results-of-artificial-intelligence-without-a-critical-view-, 01/17/2019
 Melanie Mitchell, "Artificial Intelligence Hits the Barrier of Meaning", https://www.nytimes.com/2018/11/05/opinion/artificial-intelligence-machine-learning.html, accessed on January 15th, 2019