Artificial Intelligence (AI) is an exciting topic in business, but often the vision for AI is either too futuristic or expensive to mean much to a midsized business. You might want AI capabilities, but you don’t have data scientists or high-end software developers on staff. The cool thing, though, is that AI is now becoming available as part of standard software solutions. Here are five examples of AI at work in business today.
Understanding what AI really means today
Movies and science fiction books have given us a distorted view of AI. If you think that AI can only be a machine with human-like intellect, a true Turing Test winner, if you will, then you’re missing a lot of AI’s current capabilities. (Alan Turing (1912-1954), considered one of the founders of modern computing, developed a test to see if a computer could emulate human thinking and language to the point where a person would not know he was talking to a computer.) Admittedly, very few computers can beat grand masters at chess or compose original poetry. Instead, think of AI as any software that enables a computer to make limited “thoughts” before taking an action. The following examples highlight what this looks like today.
Intelligent conversations
Chatbot software can be configured to have relatively intelligent conversations with your customers or online prospects. These chatbots won’t pass a Turing Test. However, they can be trained to ask what’s on the customer’s mind and then find answers to questions or respond to comments by searching relevant data sets.
For example, a chatbot at a retail ecommerce site could field questions on the availability of an item. If the item is out of stock, the chatbot could suggest an alternative, based on data about the person who’s asking, what’s in inventory and so forth. The idea here is to handle a question using an AI-driven chatbot rather than tying up a person on the phone. Human employees can concentrate on what humans do best and leave the repetitive, simple questions to a machine.
Automating Recruiting
The labor-intensive process of reviewing job applications and resumes can be sped up with automated, AI-driven software. Businesses are now putting this capability to work. In some cases, it’s embedded into recruitment management software packages. The software can discover qualified job candidates who were previously overlooked.
Accounting and financial management
AI is enabling software to take on tedious, repetitive tasks like routing invoices to the right approver. Now, with Robotic Process Automation (RPA), an AI-powered software “robot” (or “agent”) can perform a sequence of tasks like opening an email, reviewing the attachment and deciding if it’s an invoice. If it’s an invoice, the RPA agent can look up the likely approver from a database of purchase orders. Then, it can send the invoice to that person for approval. This saves time and frees a human accounting employee to do more productive work.
Social media analysis
When AI meets “big data,” your business can gain valuable insights into customer’s mindsets. For example, an AI-driven sentiment analysis tool can look at huge volumes of social media postings that reference your company. The tool can interpret the sentiments of the people making the posts. Then, with Machine Learning (ML), it can get better and better at spotting important trends and findings.
Fraud detection
Okay, this one is more “big company” oriented, but this kind of functionality is available to smaller firms through their banks. Banks now use AI to detect fraud. The fraud detection software uses AI-driven algorithms to review your banking transaction and “think” about whether there is suspicious activity on your account.
To do this, the software is not following a simple logical rule set. Rather, the software looks at the transaction and automatically looks at a number of independent factors to make a determination about potential fraud. It will weigh the time of day, the location of the transaction, the amount, the type of vendor and so forth to come up with a unique answer for that specific transaction.
If someone has stolen your credit card to buy a motorcycle helmet 500 miles from your home, the software will “notice” that it doesn’t look right. You don’t own a motorcycle—or at least, the software has never seen you buy anything motorcycle-related ever before. The software “knows” you aren’t on a trip. You haven’t made any travel purchases recently. It flags the transaction. That AI at work in real life.
AI is here. It’s not just for the movies or corporates with deep pockets. In some cases, AI is a feature in software you may already be using. Other times, it may take some focus and investment to put AI to work. But, in many cases, it’s an attainable technology to implement in your business.
