Artificial Intelligence (AI)

Introduction

Before explaining AI, you must understand human intelligence and how the brain works so that you can replicate the important parts. For example, understanding how the cortical columns work; they are the basic building blocks of the brain function, ie

"...cortical columns are a crucial part of the neocortex, the part of the brain that handles sight, hearing, touch, language and reason..."

Cade Metz 2018

Neuroscientists do not agree on how the neocortex works. One scientist (Jeff Hawkins) suggests that they handle every task in the same way, a sort of computer algorithm that is repeated over and over again.

One of the challenges is to duplicate the brain's capabilities.

History

In the early 1960s artificial intelligence (AI) was born with computers beating humans at chess and much progress in areas like algebra and language translation. In the 1970s, investors lost interest and this period has been known as the "AI winter".

In 2017 there was renewed interest with around 10% of all venture capital funding going to AI, ie

"...10 times more than other fields, such as block chain or virtual reality..." 

Atomica as quoted by James Titcomb 2018 

Tech giants like Google and Microsoft are focusing their companies on AI. For example, driverless cars, etc. This is building on the breakthroughs of the 50s and 60s. For example, Google's AI subsidiary DeepMind beat the world champion at Go, an ancient Chinese boardgame, that is far more complicated than chess.

Several trends are encouraging the renewed interest

- increase in number-crunching power by faster and more advanced processes 

remote cloud computing 

an explosion in the amount of data available (collected from millions of smart phones, digitalisation of records, etc) 

rise of digital games 

- interest in health care like diagnosing diseases 

The holy grail of AI is machines that learn!!! 

Need to understand AI's limitations and decide where its application is appropriate. 

Some limitations of AI include

garage in, garage out, ie 

"...if the data goes in is not complete, inaccurate or reflects biases, the decision......will not be right..."

Emma Martino-Truswell as quoted by Sally Patten 2019 

relationship becomes depersonalised, eg if using AI too much in dealing with stakeholders; this can damage an organisation's reputation. 

AI may be inappropriate in certain areas like in the last stages of hiring staff, settling employment disputes and granting of sick leave 

AI will not come up with novel solutions as it is trained on a series of examples 

One of the challenges of AI is to jump the divide from lines of code, ie following instructions, to lines of thought, ie understanding. AI is currently about lines of code while the brain involves lines of thought.

The brain is made up of billions of individual cells, eg neurons, which on their own cannot achieve much but by interacting and integrating they perform amazing tasks like a colony of ants.

 

Search For Answers

designed by: bluetinweb

We use cookies to provide you with a better service.
By continuing to use our site, you are agreeing to the use of cookies as set in our policy. I understand