Think for yourself, use properly the props of experts and artificial intelligence! But wait a minute, isn’t an expert AI supposed to help us take better decisions? As another famous HBR Press title states, “Prediction Machines” should augment our processes and help us get better products.
Well, not so quickly. In his new book, Vikram Mansharamani takes a closer look at how we abandon ourselves to the AI systems. And for good reasons – both the above mentioned titles issue a stark warning with regards to the current state of machine learning. We are in an age of great machine learning developments. And the name of this age is still infancy, therefore users beware – things have a long way to go before they get better.
Here both “Prediction Machines” and “Think for Yourself” converge on the antidote to th orchestrated jumping in. Users who led augment the AI with their own experience – after all, the human brain is likely the most advanced processing mainframe on Earth. It does not matter really how many lines of code or how many programmers worked on it – take a break, take a deep breath and think: “What is this AI product output trying to sell me?”.
As usually in our HBR books reviews on doitinvest.com, I would like to expand on a chapter I liked very much. For me, this was “Unintended blowback” – with some interesting medical cases and their impact on our lives. (This links well with the personal experiences of most friends and acquaintances). Medical science is well advanced on its way, but still has miles to go. For example rare diseases – there are too few doctors specialized in them and too few research was spent on it. As a consequence, most diagnostics for rare diseases are delivered on a qualifying manner – i.e. with a disclaimer at the end. The issues start app waring when doctors over or under interpret the diagnostics of such a rare disease. It is a bit like over or under-reacting to a serious circumstance.
We are all countries IMG in statistics and using them. What happens when you are on the wrong side of a medical statistic? You want to fight it of course – and Vikram Mansharamani makes a very nice case about it. We should use advanced AI medical software because also as patients and read the fine print on probabilities. Only this way we can improve the accuracy and advance medical science for the greater good – when we actively react and contribute ourselves, helping the experts. And not by delegating the job entirely to them.
Much has been written about expert power and how it is being used by the incumbents (see for example “Persuasion” by Roberto Cialdini. “Think for Yourself” wastes no time in explaining the basics – it just starts to make compelling practical cases for leveraging it.
Now, „Thing for Yourself“ goes much beyond that. It explores some latest research on the cognitive sciences, research which shows where we humans constantly get it wrong and where we get it right versus the machine learning. All in all, a very topical lecture for the data scientists and those interested in the latest status of using AI in business.
Title: „Think for Yourself – Restoring Common Sense in the Age of Experts and Artificial Intelligence“
Author: Vikram Manshamarani
Publishing House: Harvard Business Review Press
Publishing date: June 2020