More than three-quarters of U.S. companies are investing in artificial intelligence (AI). But according to a recent Harvard Business Review article, far too much of the investment may be going toward data scientists who can research and develop algorithms.
The majority of companies are investing in AI.
That’s not to say that data scientists aren’t of critical importance, of course. They are a bedrock upon which new AI innovation rests. And the marketplace is aware of their importance, competing for Ph.D.s in the field.
Recruiting AI Engineers Is a Top Need
But it’s also highly important, the HBR suggests, to recruit for AI engineers. People in these positions understand product development and how AI fits in, rather than understanding AI primarily and having little understanding of product development.
These roles are highly important, but lesser-filled than the data scientist role. Why? Because companies may not be focusing on the need for AI engineers.
Companies are, after all, in the business of creating products and product applications, rather than AI for AI’s sake. An AI engineer can make sure that teams develop and retain a focus on the products and how AI can serve them.
A top-quality AI engineer is equally skilled at the realms of information technology (IT), including scaling, developing software, integrating capabilities, programming, and what the needs and landscape of an organizational IT department are.
Data scientists, while important, are not the only roles needed.
Where AI Meets Business Needs and Strategy
Once AI is no longer cutting-edge research, it is likely to be commoditized. Higher-level data scientists may be less needed, and AI engineers even more key.
Ultimately, the HBR notes that the head of AI departments might be an AI engineer rather than rising from the ranks of the scientists. Once businesses develop AI sufficiently, they will be looking for a return on investment (ROI) there, just as much as from any other area. AI engineers, because of their capabilities in finding, creating, and managing data, are more likely to be helpful in the creation of ROI than many other roles.
In addition, they can be resources in being cognizant of the data sources that can help with an AI problem, can assess data quality, can clean and treat data, and can know the ways in which data can be changed.
There’s one more way, of course, in which an experienced AI engineer can be helpful. It’s where business leadership and strategy meets AI. AI isn’t just a process of developing algorithms and pushing boundaries; it’s being able to know how AI can meet the strategic goals of a company. That person can either report to the C-suite or be part of it.
Until organizations conceptualize that such a role is crucial and needed in their environment, they are not likely to be looking for such a role. And until the roles are filled, AI in business isn’t likely to return the maximum on ROI or meet its potential.