Artificial Intelligence: All Hype Aside, Let’s Get to Work

Josh Pack - Keller Schroeder Principal Consultant Data StrategyJosh Pack, PhD
Principal Consultant, Data Strategy Group

We’re not entering another “AI Winter” – those historical periods where progress is negligible, skepticism is high, and investors run for the hills.  The past decade has shown way too much technological progress and commercial deployment for artificial intelligence to go belly-up.  On the contrary, it is now a huge part of our lives in many ubiquitous ways – helping us pick our entertainment, shop for birthday gifts, and ask voice assistants for the weather.  It’s here to stay.

But let’s be frank – it’s not living up to the hype.  Autonomous vehicle commercialization has made a few modest strides, but any vision of a hands-off-the-wheel mainstream driving public is still distant, or at the very least, a niche market.  Professions dealing in repetitive cognitive tasks – lawyers, radiologists, quality control – are still in demand.  And those voice assistants, while certainly handy, are not all they’re cracked up to be.  For more on these sobering realities, check out Technology Quarterly in the June 13th issue of The Economist.

But that’s not what this article is about.  Well before “Covid Fatigue” the world began to see “AI Fatigue” (  Don’t succumb to it – that festering skepticism that sounds like one or more of the following:

  • “That’s just for the internet giants.”
  • “We don’t have the budget for it.”
  • “We’ll never find the talent for it.”
  • “There’s no business case for it.”
  • “I just don’t understand it.”

Narrow AI

Let’s demystify AI.  Autonomous vehicles, human language tasks, (plus C-3PO, Terminator’s Skynet, etc…) are all examples of “General AI,” the ability to perform a broad set of cognitive tasks and respond to the world like a rational human.  But that’s not where we’re going to improve our businesses and create the most value in our economies.  Where we need to focus is on “Narrow AI,” focused on specific actionable opportunities such as:

  • routine manual data processing in spreadsheets,
  • selecting and approving the best loan applications,
  • identifying pumps that will probably fail within a month,
  • picking the best van for a trip, or
  • detecting bad parts coming off a production line.

These are real opportunities with real value for real businesses – you don’t have to be Bezos, Ma, or Musk to set a clear strategy, make a plan, and work the plan to pursue these projects.  But you DO need to address the following:

  • Culture. This is the biggest barrier to AI – organizational silos, rigid processes, and human disincentives for innovation are the nemesis of getting AI efforts off the ground.  Change all that.
  • Capability. There’s no way around this – you have to examine your current capabilities in key areas of IT, such as infrastructure, applications development, and cybersecurity.  If these areas are too overburdened or legalistic, you do not have a fertile environment for AI initiatives.  Address it.
  • Creativity. Not everyone is meant to be a data scientist or enterprise architect, and you can’t expect digital transformation overnight.  But you do need to invest in a formal data organization that has a clear mandate, accountability, and executive sponsorship to get a firm handle on the data chaos in your organization and to convene the conversations to unlock the value lying dormant on your servers.  This is about art and science, right and left brain, vision and execution.  You can start immediately by pulling together current staff with the right mindset and appointing a leader to explore the possibilities for your organization.  Then, follow that up by figuring out who an eventual Chief Data Officer would report to (see “The Chief Data Officer Handbook for Data Governance,” by Sunil Soares).
  • Cash. Jumpstart the investment by identifying fast-payback, self-funding data projects.  Appoint and support a champion to seek and find the low-hanging fruit and compile the opportunities into a “Use Case Portfolio.”  Build relationships outside the organization, such as technology vendors, consultants, and peers in other industries, who have lived through the data strategy investment phase.  These contacts will be invaluable in guiding you to tangible returns on your data strategy journey.

So let’s not fret about AI Winters or get bogged down in AI Fatigue.  Contact Keller Schroeder’s Data Strategy Group, and let’s roll up our sleeves and get to work.


Join Our Mailing List

More Posts