The brain is an incredible instrument. It’s vital, vast and byzantine: The cerebral cortex alone features up to 33 billion neurons, and its primary physiological purpose is to exert centralized control over all other organs present in the body. This type and level of control allows instant and coordinated reactions to external stimuli. This is why the field of artificial intelligence (AI)—through which machines essentially mimic functions associated with human cognition, such as the acquisition of knowledge—is so tantalizing.
But how close, or how far away, is this promised land? And what effect will it have on us as individuals, not to mention fields such as banking?
Meet Einstein—specifically Salesforce Einstein, which just hit the market and resides at the core of the Salesforce Platform. Billed as AI for everyone, the company says the new release “delivers advanced AI capabilities to sales, service, and marketing—and enables anyone to use clicks or code to build AI-powered apps that get smarter with every interaction.”
This is exactly what makes the prospect of AI so tantalizing, and why it really does have the potential to revolutionize many common business practices. We think of digital technologies as being dynamic, and the overall field certainly is, but there’s one big exception. The apps we download and use freely keep the properties and features they have when they’re downloaded, at least until they get an upgrade.
Imagine if apps were instead more like us as people. We learn from our mistakes (or at least try to), and we act differently from the past based on changes in our environment, along with evolving tastes and needs. What if the tools we embed in, say, our mobile devices could be as flexible?
This is not as outlandish as it sounds, and some AI-related technologies are already at the foundation of many programs we use daily. Think of the way music programs like Spotify curate playlists for each customer, or Apple’s Siri builds on a combination of past behavior and geo-location to serve up choices for an evening’s entertainment, and even how Facebook’s facial recognition capabilities are uncannily accurate. This all stems from artificial intelligence.
But are we moving fast enough? Given the potential, shouldn’t there be far more AI technologies with widespread adoption?
As we’ve covered here before the possibilities are virtually limitless. The program AlphaGo taught itself to play the game Go by studying the intricacies of 100,000 matches, repeatedly playing against itself and getting better with each round. Inevitably, it beat the world’s best player. Experts predicted it would take a decade of development for this to happen, but it took barely two years.
The discipline arouses strong opposition, or at least ambivalence. Prime among critics is SpaceX founder Elon Musk—widely respected as an industry visionary, he has been quite vocal, if not always clear, on the prospects of AI. He’s backed OpenAI, a non-profit active in this discipline, but has also called it the ‘biggest existential threat’ to humanity.
Musk recently took time out to talk about AI, musing about the prospect of “improving the neural link between your cortex and your digital extension of yourself,” which would effectively turn each of us into an “AI-human symbiote.” This way, he predicted, we don’t have to worry about AI becoming an evil force, because we’ll have the same properties and capabilities.
If that sounds a little (or very) out there, it’s good to remember that this is not your average science fiction fanboy. Musk’s past interests have led to such disruptive startups-turned-giants as Tesla and PayPal, and even SpaceX—which unfortunately just suffered the explosion of a rocket two days before its launch—is by any definition an exciting venture. And sure, there’s some fanciful theorizing going on here, but every technology capability we now take for granted once seemed unlikely, even ridiculous. Is that the case here?
We know what AI can bring to board games, Siri and Salesforce tools. We know that despite conflicting sentiments, many AI startups are getting financial backing and market interest. And most of all, we know that a new generation of banking apps that are highly customizable and dynamic—learning from our habits and actions on a daily basis, reacting accordingly to a changing environment—would be truly exciting.
How would these tools drive our actions? Would they decide unilaterally which bills should be paid first, and push through payments without the user’s approval or even knowledge? Would they make recommendations on loans and mortgages? Would they choose the best investment vehicles for the short and long term? Bottom line: Would they think for us, and would that be a good thing?
The answers may be out there. But it would be better if we came up with some of our own first.