A lot of talk in digital transformation circles, not to mention business process and decision management circles, is about machine learning - Artificial Intelligence.? The words are sprinkled into conversation like magic invocations, but it turns out that AI has a long history of practical application.
- Search was long considered an AI topic
- Rules engines are a former AI topic
- Image recognition (and many other types of pattern recognition) is a former AI topic.
The old joke was, it's AI until it works, and then you call it something specific that describes what it does for you.
This article - The Singularity is Further than it Appears - has held up well over the two years since it was originally published.? It's basic argument against the "Singularity" - machine intelligence becoming self aware - is :
"Not anytime soon. Lack of incentives means very little strong AI work is happening. And even if we did develop one, it?s unlikely to have a hard takeoff."
The basic issue is that no one has the economic incentives to write something that has the potential to be emergent. But he then circles back to the full explanation, including excellent references to Vernor Vinge.
I?ve bolded that last quote because it?s key. Vinge envisions a situation where the first smarter-than-human intelligence can make an?even smarter entity?in?less time?than it took to create itself. And that this keeps continuing, at each stage, with each iteration growing shorter, until we?re down to AIs that are so hyper-intelligent that they make even smarter versions of themselves in less than a second, or less than a millisecond, or less than a microsecond, or whatever tiny fraction of time you want.
But then he goes on to refute that idea of a hard takeoff quite nicely.? It's a fantastic read.
It is also a reminder that rather than get caught up in the hype around Artificial Intelligence, we should probably focus on the benefits of more practical elements under that umbrella:
- image recognition
- sentiment analysis
- Fraud detection
- machine learning as applied to any number of business decisions with supervised training
The author even makes my point about Search:
"The most successful and profitable AI in the world is almost certainly Google Search. In fact, in Search alone, Google uses a great many AI techniques."
And he points out that the many techniques used are not about to become sentient.
So our goal with cognitive computing and machine learning is to combine it with real world process and decision management software and techniques to bring value to your business.? And that doesn't require worrying about the singularity at all.