Reading time: 4:15 

According to CEOs, only 5% of US enterprises are investing in serious AI-based business projects. CEOs are more focused on potentially existential threats from others on the Internet, firms like Amazon, leaving AI for later attention, perhaps in 2024.


Too many large enterprises have been exposed to too much market hype about AI. Too many technologists have swooned over AI’s new promises. There have been too many announcements about planned (big budget) transformative initiatives; too little evidence they ever delivered significant business outcomes.

From an enterprise executive point of view, the problem is deeper: most AI breakthroughs are technology giblets.

  • A trinket is a bauble, an ornament, a trifle of little or no worth. Giblets have value.
  • Leading-edge AI technologies are generally giblets, not trinkets…
  • CEOs do not need either technology trinkets or giblets.

What’s a giblet?

From a culinary perspective, giblet is a term for the edible offal of a fowl, typically including the heart, gizzard, liver, and other visceral organs. They’re pieces that don’t make a complete meal.

Business and technology perspective: giblets are essential components best kept out of sight. Not only are most AI technologies giblets, but, more broadly, most technologies are giblets.

Between 2013 and 2016, 49 percent of AI-related patent submissions were in computer vision technology. Giblets. (Amazing technology developments but still giblets.)

MIT Technology Review (MIT-TR) analyzed the 16,625 academic research papers on AI posted to the arXiv pre-publication server between 1993 and late 2018. They found that machine learning went from 6 percent of submitted papers in 2013 to 51 percent in 2018. That’s where the bulk of the research action is. Giblets.

MIT-TR also says

Our study of 25 years of artificial-intelligence research suggests the era of deep learning may come to an end.

Top selling giblets come and go. Attention moves to the newest, most amazing giblets. Older giblets lose the market’s attention because we learn their weaknesses and limitations.

Most CEOs aren’t going to worry about technology innovation cycles outside of their core business. You shouldn’t either, at least not until they are part of delivering a major new transformative business initiative.

CBInsights Artificial Intelligence Trends, 2019 spends 84 pages going over all the major product categories in their AI taxonomy. None of what they exhibit rises to the level of a full-blown solution that most CEOs would say constitutes a major transformative business initiative. Giblets. Many quite like offal. (It’s hard to see most robotic process automation as artificial intelligence but it’s categorized there in many market reports.)

You can include AI giblets as some of the items on a major business menu, but giblets make neither a complete meal nor a major business initiative.

Machine learning solves simple problems, but it is not sentient. And it struggles when applied to many business problems.

Source: Susan Athey, the Economics of Technology Professor at Stanford Graduate School of Business.

AI technology available today is (sometimes) amazing technology, not great solutions

Every day, we use products created or improved via breakthrough AI technologies:

  • Google Image Search
  • Chatbots and speech-to-text services from Apple Siri, Google Assistant and Amazon Echo (Alexa)
  • Automatic word and phrase suggestions while writing texts and emails
  • Automatic identification of related material when reading (as in EverNote)
  • Netflix recommendations
  • Adobe Lightroom content-aware fill
  • Facebook language translation and person identification in photos
  • GPS navigation dynamic rerouting
  • Automatically identifying plants in photos (as in Plantsnap)

This list could go on for many, many pages! We use all these tools and many, many more. We also learn that these technologies are inscrutable and only right some of the time.

On the other hand, some of what AI finally enables are game changers. For example:

  • Improving crop yields while reducing operating cost and cutting environmental impacts from over-fertilization;
  • Improving radiologists’ accuracy (and productivity) by drawing their attention to specific imaging features

Do these uses rise to the level of a CEO’s major business initiatives?

Consider Delta Airlines’ use of facial recognition technology in airports. Delta estimates that passengers could save nine minutes per flight with the new technology which, for example, eliminates the need for boarding passes.

Sounds like a good idea (provided you’re not bothered by the privacy issues that might result — participation is optional). But is it a major strategic business initiative?

The internet revolution will have a bigger impact than AI

Or so said a majority of US CEOs in the Price Waterhouse Coopers (PwC) global CEO survey discussed in my last post.

That’s telling. Is it true? In the future, the internet and AI will be so intertwined as to be indistinguishable — and throw in “digital business” into that mix as well.

In company after company, industry after industry, CEOs are seeing very serious actual or potential damage to long-standing market leaders from firms using the Internet to develop new business models, scale rapidly and dynamically adjust to a new business (and social, governmental and economic) world.

A lot of pundits, analyst firms, books and periodicals are talking about the need for “digital business” — a confusing term because so many things can be digitalized without solving the core business problem: the need to reinvent the business, change its fundamental assumptions, move to a far more dynamic, adaptive (and often highly risky) new way of working that happens to exploit digital technologies.

But just digitizing everything is too much of a focus on technical attributes and artifacts (giblets again) and too little of a focus on the new business initiatives and the means by which entities will pursue them.

Of course, digital will be everywhere (isn’t it already?)

And little (and not so little) pieces of AI will be everywhere. So will CMOS semiconductors, switching power supplies and copper wire, along with databases, advanced analytics, social computing and selfies.

Concentrating on the “digital” aspect misses the core target CEOs are faced with today: how to survive and thrive in the new world of Amazon and the few dozen other extremely aggressive competitors who are an existential threat to their firms.

In my next post, I’m going to sketch out key attributes of competitors like Amazon and identify potential characteristics of a platform that might be well coupled to CEO needs.

In the meantime, tell me what you think that platform should be like!

 

Disclaimer: This post is my own opinion. It’s not a paid post. I wrote it myself and I have no affiliation with any of the entities mentioned above.