AI hype: Primeval posturing

Describing an emerging technology’s potential by referencing precedent can be a useful way of communicating to fellow stakeholders the scope of commitment that will be needed to make an investment or ecosystem work. Naturally, plans related to a domain as pervasive and complex as artificial intelligence (AI) could benefit from this kind of framing, but one must know where the blackboarding ends and the marketing begins.

It is difficult to imagine that any technology has ever been subject to the supreme level of hyperbole that AI has seen in the past few years. When commercially viable applications began coming into focus around 10 years ago, “the new internet” was the mantra of choice. This quite reasonably indicated that AI could not only penetrate and reshape any industry, but it could also foster new business concepts.

It was a flexible representation of AI that provided an intuitive way of finding a common language to discuss how the technology could be exploited by investors, as well as how pitfalls analogous to the internet could be avoided. Now, that’s old hat.

Andrew Ng, the founder of US-based AI Fund, claims to have coined the expression, “AI is the new electricity” about five years ago. This is easy to believe considering Ng was the architect of the AI units at Google and Baidu. What’s less believable is that his wordplay may end up as influential as his industrialism. In an oft-quoted interview with Fortune, Sundar Pichai, CEO of Google’s parent company Alphabet, called AI the most important project humanity would ever work on, “more profound than fire.”

It is worth noting at this point that neither fire nor electricity are inventions – they are discoveries that have been controlled. The internet is an invention, which probably jives more closely with AI in a conceptual sense.

This is perhaps not a trivial point when one is looking for guidance on how an industry is going to develop. Will the march of AI be a matter of taming a natural resource to the point where it can be distributed and even artificially recreated? Or, will it advance more like a multifaceted discipline of computer programing?

The most bullish AI commentators will insist that the parallels with fire and electricity ring true. Problem solving, pattern recognition, and the processing of external inputs are natural processes of the mind that are now being synthesized, amplified, and packaged for the first time. As with the mastery of electricity and fire, this process is creating extraordinary opportunities to overhaul the world we live in, they may claim. But haven’t we already been here quite recently?

The algorithms and protocols that make AI tick are maturing in an unwieldy global laboratory that combines security-sensitive government operations, academic envelop pushing, and, increasingly, the commercially driven business world. This is, of course, the story of the internet, but it is by no means exclusively so; the history of nuclear energy entails similar themes. The problem is that there is only so much business-level relevance that can be squeezed out of grandiose, dawn-of-a-new-age theorizing.

Truth be told, investment and valuations in AI have accelerated far faster than technical capacity and practical deployment. A correction is therefore expected, but not a crash. After all, this is the stuff reshaping the world we live in. How can one deny its long-term bankability?

Then again, the wreckage on the road to an electrified world is no longer visible, but the wreckage of the early internet still is. Investors may do well to leverage that facility of perspective when AI excitement starts taking a viewpoint that spans the centuries.

• Tweet  
• Facebook  
• LinkedIn  
• Google plus  
• Save this article  
• Send to  

• Topics
• Technology
• artificial intelligence