Without a doubt, artificial intelligence is causing disruption to every industry, company, and geography, and its impact will be greater than that generated by the internet. Every company has its opportunity to adopt A.I. to increase revenue and decrease costs. But, this is not a time of choice. Like the internet, A.I. is an imperative. A time will come -- perhaps by 2030 or perhaps sooner -- when every company will see A.I. spread across its entire organization. Those companies that are not planning for their transformation put themselves at an existential risk.

Gartner projects that by 2021, A.I. will create $2.9 trillion in business value and generate 6.2 billion hours in worker productivity. In light of these assumptions, it's no wonder that PwC forecasts that A.I. could contribute $15.7 trillion to the global economy by 2030.

Though A.I. presents tremendous opportunity to businesses, companies will not accrue these benefits without effort. Executive teams must face A.I. with deliberate intent, realizing that in the race to gain competitive advantage from the technology, there will be winners and losers. In Notes from the AI frontier: Modeling the impact of AI on the world economy, experts from McKinsey & Company suggest that by 2030, A.I. technologies could lead to a substantial performance gap between A.I. front-runners (who fully absorb A.I. tools across their enterprises) and non-adopters or partial adopters. They quantify the rewards flowing to front-runners as a potential doubling of cash flow by 2020, implying annual net cash-flow growth of about 6 percent for the next decade-plus. This growth is at the expense of non-adopters who "might experience around a 20 percent decline in their cash flow from today's levels."

So, what can companies do today to ensure they develop a winning A.I. strategy? Here, a look into the challenges A.I. presents and how organizations can navigate these challenges to come out on top.

The Challenges Hindering A.I. Optimization

Right now, companies face multiple obstacles when it comes to capitalizing on the enormous growth opportunities A.I. presents.

For starters, there's a tremendous shortage of data scientists today, and this will remain the case for years ahead. According to job site Indeed, demand for data scientists has increased by 344 percent from 2013 to 2019, and by 29 percent in the past year alone. Meanwhile, data scientist talent grew by just 14 percent in 2018. This supply and demand imbalance means that, most often, only companies with the biggest budgets have access to data scientists.

Beyond that problem, companies wanting to adopt A.I. face internal challenges: There are up to 12 disjointed languages, seven disjointed data science tools, and five disjointed technical and business teams in an organization today. This disconnect makes it incredibly difficult to scale A.I.

Additionally, the majority of A.I. models today fail to ever make it into production, meaning many companies have not yet been able to derive massive value from A.I. initiatives. While executive teams will not deliberately allow their companies to drift to a position of losing one-fifth of their cash flow, some may be tempted to delay A.I. adoption to determine how other companies address these challenges.

However, a freewheeling wait-and-see period is an unlikely luxury--and organizations that stall may never be able to catch up to others that have successfully leveraged the opportunities presented through A.I.

Keys to Becoming an A.I. Winner

Companies planning to adopt A.I. should consider the following to become an A.I. winner and A.I.-driven enterprise:

  1. Pragmatic A.I. education for the entire organization. This should include a deep dive into what makes for a successful A.I. use cases versus those that are unsuccessful, how to overcome obstacles to success, and how to speed the path from concept to value.
  2. Value-focused A.I. Each project your company undertakes should create measurable impact to earnings, revenue, and customer experience. Rank potential use cases based on both feasibility and value.
  3. A.I. you can trust. A.I. needs to be human-friendly, intuitive and easily understandable. Companies must establish guardrails to protect themselves from inadvertent bias or errors and ensure that A.I. is working as it's expected to. Enabling technology must produce A.I. that everyone in your organization can explain and defend, and that is consistent with your ethics and values.
  4. Value-focused, agile data scientists. If your organization has data scientists, try to make them even more productive by automating as many of the manual, repetitive tasks as possible so they can focus on delivering the highest value.
  5. Capable business analysts enabled as citizen data scientists. Because the demand for A.I. exceeds the capacity of most dedicated data science teams, organizations need to identify employees who can fill the talent gap. Capable and motivated business analysts who are trained, enabled, and empowered to help build A.I. can help democratize data science.
  6. Accelerated path to production with monitoring and governance. Because machine learning models drift over time and data changes quickly, a centralized machine learning operations (MLOps) and governance center as a complement to traditional DevOps is becoming more critical to helping get a greater number models into deployment.
  7. AI transformation strategy. Finally, since A.I. will continue to be critical to business success for many years to come, your organization should start thinking now about how to inject A.I. into your corporate DNA and leverage A.I. strategically to transform your organization. This means changing your corporate DNA so that creating and working with A.I. is second nature to everyone in your company.

Becoming an A.I.-Driven Enterprise

Being A.I.-driven means taking thousands of optimized decisions collected automatically every day and using them to efficiently run your business and provide you with a strong competitive advantage. As the experts from McKinsey observe, "The size of benefits for those who move early into these technologies will build up in later years at the expense of firms with limited or no adoption."

Identify all opportunities across your enterprise to measurably impact earnings or revenue with A.I. Execute on as many of these opportunities as are justified by the business value and practicality. By following this approach and equipping your teams with an A.I.-enabled platform, your company will infuse A.I. into its corporate culture and position itself to capture the additional annual net cash-flow growth that is available only to the A.I. winners.