AI-First Digital Transformation: Opportunities & Challenges for Incumbents
“The revolution in deep nets has been very profound,” said Sergey Brin, Google co-founder, in an interview at Davos, “it definitely surprised me, even though I was sitting right there.” He is not alone in seeing the impact of AI as both profound and fast-developing.
A report by CB Insights, in 2017, showed an exponential growth in the use of AI-related terms in corporate earning calls. I think, this should be seen as a sign of executive support for AI strategies (even if, for some companies, it is more theoretical or marketing-related, and in its early days). An even more interesting story, however, is a recent chart from CBI that shows the more frequent use of “Machine Learning” than “Data Science” in earning calls. This could be an early indicator that a more scientific narrative (instead of a marketing or consulting narrative) is driving the agenda in the commercial use of AI today.
In this post, I’ll try to outline some of the opportunities and challenges for incumbents in industry verticals, such as insurance, that might be going through (or, planning to go through) their digital transformation, while showing interest in employing AI. Given the importance of AI in industries such as financial services (who are seen among top adopters of AI, according to a McKinsey report), I hope that this can help reduce the overall cost and failure risk of such strategic initiatives.
AI (unfortunately) means different things to different people
A new research by London-based investment firm, MMC Ventures, could not find any evidence of artificial intelligence applications at ~40% out of thousands of AI startups in Europe that they reviewed. Nevertheless, the companies are often described as AI-focused.
Of course, startups are not alone — many large businesses describe themselves in the same way, and are in the same boat. CEOs and marketing departments claim to have employed a strategy that can take a team of 100+ AI and digital talents to deliver, when not even a small team of modern ML (machine learning) and/or digital (e.g., design, development, product) talent is employed by the company. Unfortunately, this has resulted in AI becoming a “catch-all phrase that is used flippantly,” noted Simon Cook, chief executive of Draper Esprit.
“Almost 10% of European startups founded in 2018 were ‘AI companies’” State of AI 2019 report by MMC Ventures
This may not be a surprise to those who are observing this space. These so called “AI startups” have historically raised larger funding rounds and secured higher valuations than other software businesses. And this strategy is not limited to startups and AI. We all have seen how the stock market favours the use of buzzwords in a public company’s name/vision/strategy. Remember the story of Long Island Iced Tea Corporation?
In December 2017, The Long Island Iced Tea Corporation’s stock nearly quadrupled, after it announced that it will be “disrupting [the] lemonade category” and changing its parent company’s name to Long Blockchain Corp, with a shift in “its primary corporate focus towards the exploration of and investment in opportunities that leverage the benefits of blockchain technology.” Nearly 15 million shares exchanged hands (~50x higher than its 30-day average). This is only one example of how market rewards the use of buzzwords such as AI and blockchain.
Businesses cannot afford to see AI as an afterthought in their digital transformation
Regardless of the marketing hype, there is enough evidence to show that AI can have a profound impact on improving many (if not all) of the components of many businesses’ value-chain, as well as helping reimagine them. However, these businesses should not advocate for a digital journey that envisions AI as an afterthought.
According to some estimates, the spend in corporate digital transformation was nearly $1.5 trillion in 2018. On the other hand, research shows that 70% of complex, large-scale change programs do not reach their stated goals. For instance, Lego’s Digital Designer virtual building program, and Nike’s Nike+ Fuelband activity tracker have been defunded. Both Procter & Gamble and Burberry had ambitious programs for becoming “the most digital companies” in their industries, but ran into various challenges (e.g., growth and financial performance). And the list continues.
The AI-first digital transformation of a business — an insurance company, for instance — is the end result of reimagining every touchpoint (and, hence, the overall business model), in light of the power of machine learning to inform and assist its users, customers, and employees.
The story for employing AI will be similar (if not more expensive, slower, and with higher risk of failure as this HBR article describes). Therefore, to avoid an extreme overspending (e.g., one for a pre-AI digital transformation strategy, and one for an AI strategy to work within this new ecosystem, and additional spends on iterative improvements in the integration of the previous two strategies and making them compatible), I think that businesses should embrace AI-first digital transformation sooner — rather than later.
Let me use an example from the automobile industry to explain this. Tesla is a great example of AI-first vehicle design; the alternative can be referred to as “AI-inside”, where one adds various sensors to a “pre-AI” vehicle to enable autonomous driving (see the red car in the picture). In 2016, Tesla designed vehicles with hardware that could go all the way to SAE level five autonomous driving; they saw autonomy as a software update that will follow and hence designed the hardware for it from day one. For pre-AI vehicles, which became AI-inside, AI was clearly an afterthought; AI’s success in autonomous driving can mean that they have to go through a full redesign of their cars, manufacturing lines, and more. This is the pitfall that digital transformation initiatives in insurance (and other industries) should avoid.
This is the third article in a series of posts that I’ve decided to write, about the development of AI-first products (and AI-first digital transformation) — challenges, opportunities, and more. I would love to hear your thoughts, and any learnings and experiences you and your company might have in this space. Get in touch!
Of course, this is a personal opinion and does not necessarily reflect the viewpoints of AIG or the University of Oxford.
