In the last Part 2.1, we dove a little deeper into the meaning of AI, and explained in a simple manner how the most advanced techniques of AI (deep neural networks) were originally an inspiration of how the human brain works (biological neural networks).
In this part, we will move one step forward to highlight the similarities and differences between AI and other overlapping terms.
AI is a colloquial term that is generally used to refer to the ability of computers to stimulate intellectual tasks, and provide a human like output without being specifically programmed. The term is very much less used in the technical fields of computer and data sciences.
During a discussion initiated by computer and data scientists addressing certain AI-powered applications, terms such as machine learning, deep learning, topic modelling, knowledge representation and expert systems are more likely to be used than AI. Nevertheless, AI is still a dominant term that refers to all underlying techniques and algorithms capable of stimulating some intelligent tasks.
Some argue that when computers are capable of completely performing intelligent tasks and generating a human-like output, the task is no longer intelligent and loses the quality of being qualified as an AI-generated output. This is because the computer does it seamlessly and regularly. While the argument may seem academic to a certain extent, its practical demonstration could be very interesting. The fact that an intellectual task can be performed by a computer with no margin of error, and without being specifically programmed means that computer is likely capable of performing, or at least ready to learn, a far more intelligence new task while building on the previous one. This point will later take us to the meaning of Artificial General Intelligence.
The term AI is also confused with other terms such as Business Intelligence (BI) and such confusion is, in my opinion, one of the reasons why AI is sometimes overhyped and oversold in the business world. That is why it would be important to draw the line between the two fields and shed some light on how AI can help augment BI.
Artificial Intelligence vs. Business Intelligence
BI refers to the set of tools and technologies that are used to collect, store, access, analyse and transform data into actionable analytics in order to help businesses make decisions in a faster and more efficient manner. BI enables organisations to improve the quality of data they collect, and the consistency with which they collect it. BI tools also streamline the tasks that employees need to carry out in order to search for, collect, merge and analyse data to obtain the information they need to take sound business decisions.
One of the key benefits of BI is that the collected data can be broken down into a multitude of ways to support different roles, tasks or decisions.
Take the example of an organisation (such as a law or an auditing firm) that charges its customers based on an hourly rate. Each employee (lawyer or auditor) has a role to be performed in the form of deliverables, such as legal review of documents or providing audit reports based on their time. These employees are called fee earners. With BI tools, fee-earners can look at their own billing by matter, over time, and against their targets, and can compare their performance with others whereas the head of the department (such as a senior partner) could look at billing or other indicators for their whole department, and then break it down by sector matter, client, team or fee-earner.
Meanwhile, business support departments within the firm, such as HR and finance, could focus on different metrics according to their strategic priorities.
BI tools could also be very helpful for retail businesses with seasonal business cycles. Such businesses find it challenging to optimise their stocks to cope with seasonal demands. With BI tools of decision support systems, analysis of historical sales and stocktaking data for warehouse products, such businesses have been able to significantly increase their profitability and efficiently meet market demands.
There are many leading BI service providers such as Microsoft, Qlik, Tableau, IBM, Oracle and SAS. Organisations choose their BI platforms based on various factors depending on the size and complexity of its operations.
Nevertheless, BI is good as far as providing tools to let the organisation take the right decision. It does not make any recommendation or bridge this causal link between the analysed data and the decision to be taken.
As one of the leading decision science professors put it “BI does not tell you what to do; it tells you what was and what it is”. Simply put, by way of analogy, BI cleans and organizes your dressing room, but it does not tell you what to wear on a given occasion.
But AI is different. It can bridge that causal link between the data (even in its raw forms), and the decision to be made. AI can make the decision the way of finding the patterns in the data and applying the right algorithmic operations to such patterns. A computer trained on a hundred thousand photos of dogs and cats, can tell you whether a new photo is a cat or a dog. AI does not only cluster the historical photos of cats and dogs (like BI) but also decides what object in a new photo is.
AI is a perfect fit to help augment BI tools to offer organisations integrated platforms that do not only analyse data, but also generate human-like outputs using such data. AI is a great interface that could help businesses optimise the efficiency of their time and expenditures, as well as direct their human resources towards smarter and career-progressive roles.
Next Article Part 2.3: AI and AGI
Hani A. Rasoul Chief Executive Officer & Chief of Legal Tech. & Analytics at Brightiom