In a previous blog post, we took a look at the various forms of artificial intelligence that are in use today. These include descriptive, predictive, and prescriptive analytics. However, these are just tools. The far more impactful area, in terms of how this technology affects people’s lives, is where the brilliance of AI and analytics solutions are put to use in the marketplace.
Major companies such as Amazon and Apple have increasingly capitalized on these tools in the past five years — so much so that the NASDAQ blog referred to 2018 as The Artificial Intelligence Boom. The private sector implementation of AI can be broken down into four major categories: Internet, Perceptual, Autonomous, and Business AI. Each form of implementation finds a clear benefit in the analysis and implementation of data.
The private sector’s approach to internet AI is given a perfect test case through the comparison between Spotify and Apple Music’s curation algorithms. On a basic level, internet AI is the use of an individual’s browsing and consumption habits to aid in the suggestion process for suppliers. The most famous of these would likely be the Netflix Algorithm, which looks at user data to form “taste clusters.” Netflix produces shows and markets them to directly fit these clusters.
Spotify and Apple Music offer an interesting example because of their differing approaches to the issue. Spotify has leaned in heavily to the idea of internet AI, while Apple has remained reserved, favoring a more human-centered curation strategy. It seems the decision from Spotify is business related and Apple is optics related. The Swedish-born streaming upstart uses a streaming algorithm that is so successful because, as Robert Safian writes in a Fast Company article, “Its technology interacts with music, artists, and listeners. There are so many songs, and so many music tastes, that without cutting-edge software, it’s impossible to sort and prioritize all the options.”
On the other hand, Apple Music prefers a more hand-crafted approach with Tim Cook criticizing those who turn music into, “bits and bytes,” according to Safian. The difference between Spotify and Apple on this topic is indicative of the public backlash to AI in the sale of entertainment or art. Surely Apple Music is using an algorithm to suggest music. However, the company signals it is against it because the process comes off as a cold, big brother strategy that will only keep users within their comfort zone. Spotify, however, paints itself as a hip, up-and-coming tech company similar to Netflix. Because of this, users are not as turned off by its use of internet AI in the suggestion process. In each case, the use of internet AI is impacting what songs people listen to, but the nature of the process requires different treatment for both companies.
Perceptual AI mimics human speech and language. It has been popularized by Amazon’s Alexa, Apple’s Siri, and Microsoft’s Cortana. This technology is commonly involved in human computer interaction (HCI) and in the article “The Future of AI in Distribution,” authors Ian Heller, Michael Wu, and Austin Garrison say the solutions are able to, “‘learn’ your way of speaking over time and get better at understanding what you’re asking them to do.”
However interesting these technologies are on a superficial level, they offer a wide array of upside for the businesses listed above. Let’s tackle three specific positives that Amazon sees from Alexa and perceptual AI. Most notably, Alexa makes it easier for customers to purchase goods off of Amazon. This lends itself nicely to groceries or everyday items that may pop into a user’s head as opposed to shoes, for example, which take a little more online digging.
In addition, the perceptual AI platform excels at bundling items and integrating Amazon firmly into a user’s everyday experience. First, Alexa comes with Amazon Prime, which gets the TV connected. Then there are the other items that can be connected to the device like smart lightbulbs, thermostats, alarm systems, and kitchen appliances, which all add to the inundation of Amazon products in the household. In a similar vein, the pairing of Alexa to other items in the home opens up opportunities for partnering with other services “for the purpose of expanding the ecosystem and gaining a larger installed base,” according to the tech blog Innovation Tactics The advantages of perceptual AI are far more widely accepted than those of internet AI and are sure to propagate as the internet of things gains traction in the marketplace.
Automation is likely the most familiar and dreaded form of AI for the average American worker. This solution is simply robots physically doing human things. Some examples would be factories with robot operators, Amazon distribution centers with robot handlers, or taxis with robot drivers. The benefits companies could see from this change are clearly huge. With autonomous factory work, salaries would be replaced by maintenance costs and instances of human-error in the system would plummet, both of which would help the bottom line of the company.
But with every positive there is a negative. In this case, the negative would fall on the millions put out of work because of these changes. These fears are, however, overblown according to a McKinsey Global Institute Executive Briefing which states that by 2030 the increased demand for labor driven by automation will offset the number of jobs lost because of it. Some markets will obviously be affected more than others but, in the end, more jobs will be changed by automation compared to lost or gained. Autonomous AI, like its perceptual counterpart, is far from reaching its full potential and will dramatically alter the global labor market lead by forward thinking tech companies.
Business AI focuses on prescriptive analytics to diagnose an issue within an organization and then provide a solution. The “Future of AI in Distribution” article says business AI can be anything from fraud protection to machine-aided medical diagnosis. A more concrete example of business AI optimizing a company’s process comes from UPS which spends, “about $1 billion annually on on-road integrated optimization and navigation.” This money goes towards monitoring driver and traffic habits, recommending best routes, and predicting fleet maintenance. The savings derived from this business AI are about 100 million miles and 10 million gallons of gas annually. This form of advanced analytics works to perfect the operations of a company and improve its bottom line.
Artificial intelligence through analytics is interesting in theory, but it could change the world when implemented in the private sector. Its benefits mean far greater profits for companies like UPS and Spotify and more integrated users like those who will have Amazon’s Alexa connected to every appliance in their home. There are negatives to such a large process too, as many will lose their jobs or feel their privacy is being invaded by the selling of user data. Whether one falls closer to Apple or Spotify on the debate surrounding AI in the business world, there is no question surrounding its impact.
- The Evolution of Artificial Intelligence in Analytics
- 3 Advantages to Using Simulation in Predictive Analytics