In the movie "Her," the main character, Theodore Twombly falls in love with the operating system of his computer. The operating system is run by an artificial intelligence who Theodore calls Samantha (voiced by Scarlett Johansson), who over the course of the film develops self awareness and shows capabilities far beyond its creators’ intentions.
The film is set in the so-called near future, a recognisable Los Angeles/Shanghai hybridisation with elements of the present projected into what could possibly be. Without spoiling the film for those who haven’t seen it yet, the capabilities of the AI are what we imagine artificial intelligences to be. The science fiction aspect is that the film takes these imaginings and animates them, letting us forget the harsh reality of the present.
That reality is reflected by companies like Google and Amazon who are using algorithms in their attempts to show intelligence that is so basic as to make the future presented in her seem a very long way into the future. Take for example the current personalised advertising that Google has implemented. Have you noticed how if you happen to search for an item, all of a sudden you get deluged by adverts for that particular item or service? There is nothing subtle or sophisticated about this. Google has decided that because you searched for an item you are interested in it and in case you didn’t make a purchase, it will keep reminding you about the service at every available opportunity. Clicking on an ad gives you the same result.
It turns out that this is not simply our imaginings, it is a feature called Ad remarketing. The basis of this is that only 2% of people visiting a site will complete a purchase and so going after the 98% who didn’t gives retailers a way of going after that interested market. There is nothing subtle in this approach. It is a particularly blunt force approach to selling.
So when Google talks about using sophisticated and personalised advertising, this is as good as it gets. Forget the hype of Big Data and detailed analytics, this is a simple if-then decision. If person has visited site then show ads for site repeatedly.
There is a strong incentive for companies, organisations and individuals to show that they are innovative and represent the future. Equally, there is a huge risk associated with being labelled as a follower or un-innovative. Apple’s fortunes for example, have largely languished over the past year because of a perception that they are not able to innovate any more.
One of the main issues facing companies is the fact that the image of innovation is often set by the marketing department and is not reflected by the organisation as a whole. There is a belief that creating a few mobile phone apps is all that it takes to be innovative in the eyes of your clients and the general public. A good example of this is the Commonwealth Bank who has been lauded as being innovative. As I have written before making token efforts to appear innovative with mobile apps will utlimately fail if the entire organisation is not operating at the same level.
It takes very little interaction with the bank to realise that behind the online high-tech facade, there are administrators largely still operating manually. An application for a credit card made online for example results in letters sent by regular post. There is no AI here, not even online communication. Ultimately, you still have to talk to a person and go through the same arcane process (and arcane credit rating process) that has existed for decades.
I enjoyed the movie "Her." For a couple of hours, I was able to imagine the possibilities of technology and how as a society we will need to adapt if it were ever to eventuate. As the adaptation of humans is really the biggest challenge in this equation, it is fortunate perhaps that the future presented in her is not remotely near, even though people may like to pretend it is.
David Glance does not work for, consult to, own shares in or receive funding from any company or organisation that would benefit from this article, and has no relevant affiliations.
This article was originally published at The Conversation. Read the original article.