The science for artificial intelligence (AI) has been around for almost a century. Although the concept of AI can be traced back to ancient stories from Greece and China, AI practitioners have only actively achieved significant progress within this century. The approach to applying AI has been continuously refined, from using symbolic or classical AI that is based on embedding knowledge and creating rules, to progressing statistical AI which involves pattern recognition, probabilities, and learning. Nonetheless, the challenges encountered in applying AI have forced renewed thinking among practitioners.
With computing power increasing exponentially over the years, the discipline of AI now has an explosive opportunity to grow whatever the approach may be, but largely towards practical use cases. In addition, the growth of AI would not be centered solely around creating robots that think and behave like human beings, as is often portrayed in movies and books. Instead, it will revolve more around how it can fulfill specific objectives as agents in various forms to help humans achieve their work. This is where the current boom is coming from, but this unfortunately also raises concerns related to how AI may replace some jobs. On the other hand, there also are many who are actively pushing and exploring the possibilities for AI to make work more efficient for people.
In the recent EY Tech Horizon report, where about 1,600 companies across the globe of varying sizes were surveyed quantitatively and qualitatively, respondents reported that AI and machine learning will be a significant part of their top technology investments within the next two years. AI can be considered a foundational technology, along with data and analytics, the cloud, and the Internet of Things. By combining these four as part of their technology strategy, organizations gain a good foundation to execute a tech-enabled transformation. As AI continues to gain more traction in development and adoption, however, its prominence as an expected technology stack will likely increase.
CHALLENGES IN AI ADOPTION
The recent Forrester’s Global AI Software Forecast predicted that off-the-shelf and custom AI software spend will double from $33 billion in 2021 to $64 billion in 2025. Moreover, AI software is projected to have an annual growth rate of 18%, which is 50% faster than the overall software market. However, based on one IDC survey, only 22% of organizations in 2022 reported that AI was implemented on a large scale in their enterprises.
For all the reports around how AI will be a priority for investment and implementation in organizations, we have yet to see it leveraged at scale to transform businesses. This poses the question — why is AI, which is clearly a top-of-mind technology around the world, seemingly unable to make a strong breakthrough?
A key challenge for widely adopting AI across an enterprise is the typical bottom-up approach taken by organizations. It often starts with technical teams building portfolios of projects on piloting AI on specific processes only, which is disconnected from the wider business. This creates siloed solutioning and tends to put the burden on technical teams to foster the business cases for AI to incrementally secure more funding. This also means that solutions may not look holistically at how the business works and can lead to an incoherent approach to delivering value across functions. This incoherence can lead to leadership and teams being disillusioned with what AI promises to deliver.
Maximizing the potential of AI requires a refreshed approach. Businesses should include this as a strategy from the top to help differentiate themselves from competitors while simultaneously becoming more resilient to disruptors in their industry. Chief Executive Officers (CEOs) should drive the change in culture towards AI and align with their Chief Information Officers (CIOs) towards championing the importance of technology in meeting organizational goals.
This means that there should not be a separate technology strategy, but one embedded into the business strategy itself. This approach can help answer questions such as whether AI allows the organization to enter a new market or re-imagine its business model. This will ensure that investments in technology will align and help meet the vision of the organization.
As early as the 1970s, one of the first recognized knowledge systems (classical AI approach) was MYCIN. This system was intended as a doctor’s assistant refined over a period of five years and was designed to give input and explanations about blood diseases based on blood tests.
The use of AI has since been developing across various industries and businesses should be more aware of how it can be deliberately leveraged while being driven by an organization’s leadership. The most common use is enabling it as a recommendation engine for online retail or search, which by now should start being a standard for most online stores.
Fast forward to now: machine learning (statistical AI) can help identify diseases in medical surgery, from CT scans used to detect lung cancer to eye photographs to check for diabetes. Another example is in smart farming, where crop scanners and drones collect images and analyze them to determine if more water, pesticide, or fertilizer is needed. In short, regardless of the industry, AI can drive innovation to help humans.
AI can clearly help enable CEOs innovate, but with it comes the responsibility of proper implementation. There are issues such as data privacy, embedded biases, and accountability in deploying this technology. While the strategy is defined from the top, along with it comes the necessary governance to help safeguard the organization from its risks. In 2023, seven leading AI companies in the US have agreed to voluntarily implement safeguards on AI development. They have committed to establishing new standards for safety, security, and trust for the technology.
Being familiar with these movements can help business leadership navigate where to plan for AI adoption, and check for readiness on how to address potential regulations in their industries.
EMBRACING THE IMMINENCE OF AI
We have come to a point in history where AI can be deployed as a day-to-day technology. This was made possible because the science now focuses more on specific use cases rather than creating general AI that strives for human-like intelligence. AI is going to be an imminent part of how businesses will be run today in the same way that machines in the industrial revolution moved workers from cottage industries to industrial factories and introduced new skills for people.
CEOs must embrace this imminence and can do so by intentionally including this in their high-level strategy, understanding how the technology can be applied today, upskilling employees to work with AI, and more importantly being proactive on internal governance and external regulation.
This article is for general information only and is not a substitute for professional advice where the facts and circumstances warrant. The views and opinions expressed above are those of the author and do not necessarily represent the views of SGV & Co.
Lee Carlo B. Abadia is a technology consulting principal of SGV & Co.