Page 39 - Kaleidoscope Academic Conference Proceedings 2020

Level 3: Anticipation and identifying situations before they as commonly imagined, about thinking like a human being.
occur, providing the organizations with time to exercise For the time being we are not close to having “General AI”
remedial options. (Prognostic) that will outthink us or come close to having the qualities that
we cherish in our own capabilities and imperfection. It is
Level 4: Discovering problems and at the same time hard to predict when we will perfect the technologies that
identifying solutions or courses of action and remedies, with will achieve such a level of capability. However, we have
people in the loop to authorize. (Prescriptive) long ago passed the point where AI will outperform us in
specific ways and hence is central to delivering the value
Level 5: Improving and optimizing on a continuous basis, from digitization.
and discovering through learning alternate, and sometimes
unexpected paths to better outcomes (Autonomic) Within AI there is a class of techniques built around machine
learning (ML), where there is something profoundly new and
A recurring question I hear is: Which industrial vertical that is what I will focus on. There are three aspects to this:
markets will be affected by digital transformation? The (1) the underlying algorithms, which while essential have
simple answer is that almost all will be, maybe in different been around for some time; (2) the discovery that large
ways, but eventually the impact will be felt in almost every volumes of data can profoundly improve the quality of
sector. The thing to watch is how various industries progress results for a large classes of applications; and (3) advances
along the ‘Level’ scale in their performance in adopting in computer performance that have made some of these
digitization. The precursors are already there, and the applications practical. The second area is all about data and
evidence has slowly crept up on us, so we are not always how we go about using it, and why it is probably the biggest
conscious of how deep the toehold already is. It’s worthwhile impediment and challenge for digital transformation. The
illustrating with several examples that are indicative of the problem with data is both technical and organizational, so I
diversity involved. would like to explore how both aspects can be fixed.

The first is financial services: how we are paid, how we Gartner reports have captured the dynamic evolution of
purchase things, manage our assets, and decide on our technologies in both AI/ML and in data and their state of
investments. Today in many organizations, we report our maturity as illustrated from the Hype-Curves for 2019 and
time and effort electronically, and can perform such tasks 2020 in Figure 2.
from anywhere on an app. The funds we are owed are
transferred to our accounts electronically too. For gig-work,
we bill electronically and expect to be paid electronically no
matter where we work and no matter where we are delivering
the results. To pay for goods and everyday staples, we
increasingly use credit cards or other electronic payment
systems and often allow for recurring payments to be made
automatically from our accounts. The same trend is true for
how we budget and do accounting and how we elect to
deploy our savings. What we feel as consumers has also been
deeply felt across the financial industry, from digitization of
financial and asset records to the way business is conducted
at both the consumer level and in serving industrial
businesses all digitally. Not to say that it works well, but the
evaluation of assets, determination of creditworthiness,
granting of loans, risk analysis, and issuance of insurance is
increasingly conducted digitally. The same is true of the
trading of assets of almost any kind – artwork, real estate,
stocks and bonds, commodities, or other forms of equity.
Underlying these services is mountains of data and an
incredible range of AI and analytics tools.

4. TECHNOLOGIES IN FOCUS

I have enumerated the plethora of important technologies
that are important in the context of digital transformation.
There are two that are “Super” important on their own and at
the same time inextricably intertwined; these are the
technologies around artificial intelligence (AI) and data
science and engineering (DS&E) The first of these, AI, is
broadly speaking about the exploitation of computers to
perform tasks better than we can as human beings. It is not,

– xxxv –