Jul
31
With every Siri, Alexa, Google Assistant, and Cortana upgrade, conversations about AI—artificial intelligence—inevitably arise. Also inevitably, out roll the usual questions from definitions to ethics, usually set against the backdrop of a dystopian future. But in the here and now, the growth and status of AI-like applications portend no small developments for the financial services industry.
The above-referenced ethical questions often have to do with whether a machine with humanlike sentience should be accorded human rights. These sometimes sprout from fanciful views such as those held by computer scientist, inventor, author—and would-be futurist—Ray Kurzweil. His bona fides as an inventor and computer scientist are well established, but Kurzweil-as-futurist waxes a bit over the top. In a 2005 TED talk, Kurzweil predicted that in 2029 nano bots will …
… go inside our brain, interact with our biological neurons … shut down the signals coming from your real senses, replace them with the signals that your brain would be receiving if you were in the virtual environment … a tremendous expansion of human intelligence through this direct merger with our technology …
I kinda doubt it.
Some equate AI with a machine’s ability to pass the Turing Test. The 2014 movie The Imitation Game made the late mathematician Alan Turing a household name. Turing famously proposed a test wherein an unseen machine would exchange notes with a human. If the machine could convince the subject that it was a human, it would be reasonable to say that it was capable of imitating human thought.
There’s an important distinction to be made: Imitating human thought isn’t thinking. But there are many who argue that AI needn’t necessarily invoke an Isaac Asimov-esque threat from cyborgs or even pass the Turing Test. For that matter, AI needn’t be good at all things. It can be good at just one thing.
Take apps like the above-referenced Siri, Alexa, Google Assistant, and Cortana. They are by no means sentient, nor are they capable of reasoning, but they are great at recognizing human speech across multiple languages (according to MacRumors, Apple leads the pack with “21 languages, localized for 36 countries”) and, to a point, learning how better to interact with users. Considering regional accents and the many ways we humans can express the same thought, that’s no small feat. But then, last year Siri alone had 2 billion users requests—per week—at its disposal for analysis.
Self-driving cars are also an arguable foray into AI. They must make decisions that look like thinking—the programming this requires is beyond me—but there will likely be no self-driving car apocalypse when a Smart car suddenly lives up to its name and organizes its fellow vehicles to take over the world.
Most important for our purposes, today you can order
Alexa to help with basic banking transactions …
… which Extractable’s Chief Analytics Officer Mark Ryan happens to see as the beginning of a trend. He wrote in an article for The Financial Brand,
In the next 5 to 10 years, we will most likely see Web traffic to banking sites and mobile applications drop by over 50%. We know that an average of over 50% of the traffic to online banking through browsers and mobile applications is there to perform the simple task of checking a balance and then leaving.
As voice recognition and voice authentication mature, banks will be able to offer customers the ability to perform their most common banking tasks such as checking balances by simply talking to an Internet-connected device (i.e. car, television, watch). The integration of more voice devices (and WiFi connections) in automobiles may be one of the major tipping points.
If Ryan is right, then voice-activated banking might be the nearest thing to an AI apocalypse on the financial services horizon. But “apocalypse” is a bit melodramatic. After all, digital banking was no apocalypse. It was just one heck of an opportunity, and the banking industry is doing a good job of seizing it. I suspect we’ll do the same with voice-activated banking.