AI is seeping into enterprises from all directions. It's being embedded in applications, software tools, devices and equipment. Yet, some organizations still don't have an AI strategy.
"The journey to AI having an impact at a firm is challenging and sometimes long," said Nigel Duffy, global artificial intelligence leader at professional services firm EY. "To have an impact with AI you must solve a lot of problems, many of which have nothing to do directly with AI, [such as] how do you deploy solutions? How do you get them into your infrastructure? How do you get people to use them? What are the workforce and change management considerations? What kind of training is required? What's the governance model?"
AI is too disruptive to ignore. If your company lacks an AI strategy by design, then it has one by default. A default strategy is a Wild, Wild West scenario in which AI is popping up in various places within an organization, without orchestration and alignment. The lack of cohesion and direction can result in several issues including governance and security.
"[S]ome of those risks may not be well characterized, so they have not been addressed by the appropriate level of governance and review," said Duffy. "Most AI is going to come through procurement or it's going to come through the backdoor or technology you've deployed."
Should an AI strategy be part of a digital transformation strategy?
One reason some people think an AI strategy should be part of a digital transformation strategy is because the digital transformation strategy is viewed as the overarching business initiative that's facilitated by modern technologies.
"If an organization’s digital transformation strategy does not already include AI, then there is a real need to revisit the overall approach to transformation, said David Homa, director of the Digital Initiative at Harvard Business School. "Technology expertise and knowledge must become foundational to how an organization conceives of, not just executes, its goals. As long as technology is viewed as a tool provided by a supporting team, rather than a lens through which to understand the business and produce value, the organization will limit itself and lag others who integrate technology into the core."
Some think the AI strategy and digital transformation strategy should be separate but aligned, including Manish Mistry, CTO at digital engineering services company Infostretch.
"An enterprise’s digital transformation strategy generally is dictated by the business objectives, with targeted goals and outcomes at the core. Companies are still trying to figure out what the actual business impact of AI is, and that depends on the maturity of the company and the industry in which it is evolving," said Mistry. "A lot of businesses are still trying to figure out how to collect data, what type of data to collect and how to integrate it. Some companies are capitalizing both on their AI strategy and DX strategy, but the reality is not all of them have reached that level of maturity. They are side-by-side initiatives."
Justin Silver, data science manager and AI strategist at AI SaaS solutions provider PROS thinks AI and digital transformation strategies should be closely linked.
"Organizations are moving more quickly than ever before to digitize business processes using AI to make those processes smarter and more efficient. However, there are additional considerations for an AI strategy beyond digital transformation," said Silver. "Sometimes business leaders consider digital transformation use cases and ask, 'Can I apply AI to this?' The answer is that if you can fully digitize a process, you can apply AI to it. A more productive question to ask is, 'What’s the value proposition of applying AI to this use case?’”
Reasons why the AI strategy should drive digital transformation
EY's Duffy and Linda Giuliano, founder and CEO of environmental, social, and governance (ESG) consultancy BrightWorld Advisors argue that an organization's overarching strategy should be its AI strategy.
"AI strategy touches so many aspects of a business [including] talent, culture, innovation and employee engagement, said Giuliano. "What does that mean in terms of attracting new talent and retraining people?"
Quite often, organizations are doing AI proofs of concept (POCs) without considering the broader range of use cases or the impact AI will have throughout the organization. This is one reason why "successful" AI POCs fail to scale well. They're too narrow and they haven't been undertaken within the context of what the organization is trying to accomplish.
"[E]very strategy should be an AI strategy as opposed to having AI be a component of the strategy. Even your acquisition strategy should be informed by the impact that AI is going to have on your industry segment and your competitors," said EY's Duffy. "AI is going to drive so much change, both in the environments in which you're operating and also internally."
The idea is to adopt an AI mindset when planning and executing a digital transformation strategy, such as how it will change the nature of customer relationships which differs from simply automating an existing process.
"Whether you're designing new business models or your digital strategy or even your acquisition strategy, you need to be significantly informed by where AI is taking [you]," said Duffy. "A lot of leadership teams are probably not ready to design their firm about what's happening in the world of AI and they need to get there. Whatever strategy you're designing or implementing, whatever problem you're trying to solve, you should be asking "How can AI play a role here?' or "How does AI change the landscape?"
Digital transformation strategies conceived in the absence of an AI strategy may also fail to consider important AI ethics issues such as dubious data usage and the potential negative impacts of bias.
"I don't think this is led just by the head of IT. I think the business unit heads need to be involved in those discussions. It needs to feed up to the most senior levels of management and ultimately the board of directors in terms of oversight," said BrightWorld Advisors' Giuliano. "They need to understand the opportunities and risks."
Follow up with these articles on AI strategy: