As I shared in Part One of this series, I’ve been on a journey since early 2017 to get educated about Artificial Intelligence (AI). I was overwhelmed by the volume of articles published about AI and a proliferation of claims related to the use of AI technologies that were confusing to me. I craved a framework that would simplify and clarify what AI is all about today and how to think about taking advantage of the technologies. This framework is discussed in Part One.
In this second article, I want to share my perspective on what HR leaders need to do with this knowledge. I’m writing this article with a focus on senior HR leaders as the target audience, but I hope others can also find some of these thoughts to be useful.
I believe there are at least three major actionable implications for HR leaders as AI technology utilization advances:
HR leaders should be working with others in their companies to find opportunities to apply these technologies for the benefit of customers and employees. This is simply part of the ongoing responsibility of HR leaders to periodically step back and look at the whole system of organizational effectiveness that is integral to achieving the strategy of the company. Senior HR leaders should always be looking at core value-creation processes and infrastructure support processes to test whether the application of new technologies could improve the performance of these processes. This requires that these leaders be deeply engaged in all aspects of the business so that they can start with a clear picture of the opportunity or challenge – and then determine where these new technologies can help.
Spotting these opportunities also requires leaders to be sufficiently educated about the potential of these new technologies. I refer you to Part 1 where I provided an overview of the field and references to additional reading. There are now many online courses available to learn about AI from providers such as LinkedIn Learning and major universities.
A good example of applying AI to core value creation processes comes from PayPal. PayPal COO Bill Ready was recently interviewed by McKinsey about the application of AI to PayPal’s products. Ready provided this example:
We use a lot of machine-learning technologies in the area of fraud and risk capabilities. PayPal One Touch now has more than 70 million users. It’s become the most rapidly adopted product in the history of PayPal. We are able to give a user a seamless buying experience—no password or fingerprint required—because we’re able to consume hundreds of data elements on any one transaction and secure the user better using those elements than if the user had given us a password.
PayPal has taken one of the basic things that merchants value – getting paid – and simplifying that in a way that customers like - using AI technologies.
There are some philosophical choices that must be made about the application of these technologies to business opportunities. I’m not going to attempt to cover complicated legal and ethical considerations here (Jeffrey Kaplan does a good job of that in Artificial Intelligence: What Everyone Needs to Know). In this article, I’m focused on the question of whether to see the application of AI technologies as all about productivity through automating work and eliminating jobs or as augmenting humans and creating new opportunities for people. Here’s an example. I’ve recently talked with leaders in a financial services company about how they intend to apply these technologies to financial planning advice. They were grappling with the question of whether they wanted the technologies to completely replace financial advisors, augment financial advisors by eliminating some of the simpler tasks that were taking up their time or some hybrid of the two. When I last spoke with them, their philosophy was to augment financial advisors – but there was still much to consider about how that would work.
A lot has been written on the employment opportunities AI technologies create by eliminating monotonous work or by figuring out ways to make people super-human. Steve Lohr wrote in the New York Times recently about one such example at State Auto Insurance Companies. He quoted a manager in the AI effort who said: “We’re here to partner with you [the employees] to find those tasks that drive you crazy.” In this case, State Auto Insurance Companies applied some AI technologies in the automation of tasks and then tracked the efficiencies. The results showed that AI technologies allowed the company to increase throughput significantly without adding much headcount. These stories seem to be proliferating, creating hope for these positive outcomes in the near term.
I feel both optimism for those possibilities, but caution to understand all of the implications for a company’s workforce over the long run. It’s important that HR leaders be engaged in this debate. The debate is partly about the business strategy – but it also has to incorporate the implications for the workforce and the culture of the company.
In workforce planning, HR leaders should be looking at the jobs that will be changed the most by application of AI technologies and considering how to help prepare the workforce for those changes. This is similar to prior waves of automation and business process transformation - but it is happening at an accelerated rate.
For those early in the journey, the McKinsey Global Institute published an excellent study in 2017: “A Future That Works: Automation, Employment and Productivity.” The study provides a framework for thinking about the types of tasks that present the best opportunities for the application of current state of the art AI technologies. It breaks jobs down into 2,000 different tasks and analyzes the potential for automation. Some of their conclusions: less than 5% of occupations can be automated entirely – but about 60% of occupations have at least 30% of tasks that can be automated. The authors identify seven categories of activities that are involved in work. They note that the potential to automate using AI and other technologies is high for tasks that involve processing and collecting data and for predictable physical tasks. Tasks that involve managing people, applying expertise to creative tasks, and interfacing with stakeholders have a low percentage that can be automated.
An example from the financial services world illustrates this framework. J.P. Morgan released a study last year titled “Big Data and AI Strategies.” The study was summarized in an article by eFinancialCareers. The J.P. Morgan team found that human beings are already all but excluded from high-frequency trading and in the future, machines will increasingly take over additional tasks. They were quoted as saying: "Machines have the ability to quickly analyze news feeds and tweets, process earnings statements, scrape websites, and trade on these instantaneously. This will help erode demand for fundamental analysts, equity long-short managers, and macro investors.” At the same time, humans retain certain advantages. According to the authors, "machines will likely not do well in assessing regime changes (market turning points) and forecasts which involve interpreting more complicated human responses such as those of politicians and central bankers, understanding client positioning, or anticipating crowding." This is where investment professionals will need to focus their skills and contributions.
While automation may overtake some tasks or entire jobs, many predict that AI technologies will require the creation of new tasks and roles. The Summer 2017 issue of the MIT Sloan Management Review talks about some of these new jobs. One example is the people who have to “train” a supervised machine learning algorithm. This includes organizing the input data in a way that maximizes chances of productive results and avoids introducing bias into the algorithm and then fine-tuning the algorithm through the supervised learning process. Another example is the people who specialize in forensic analysts who have to “explain” how a system makes decisions so that humans can learn from the insights (and auditors can validate processes). In the J.P. Morgan example, the authors of the study see the need for an army of data scientists and quantitative researchers who will acquire, clean, and assess the data needed to implement machine learning technologies that can derive tradable signals and insights.
All of this requires a workforce that will focus on and improve the skills that can’t be readily automated. It is a significant challenge and investment to conduct workforce planning at a detail skill level and then develop learning plans to fill gaps that the labor market may not yet be able to fill. FLEX is a company that is currently being proactive in preparing their employees for the future. They recently wrote on their website:
There are many examples of how we have taken up the challenge to upskill our employees, including a Business Process Automation and Artificial Intelligence training designed by our Analytics and AI experts to raise basic awareness….We provide opportunities for employees to develop automations that support their own day-to-day activities and advance their professional skillset in the process. In partnership with an internal developer, employees can test and deploy their own automated office solutions on computers with Robotic Process Automation (RPA) software.
FLEX is also closely monitoring the corporate culture impact as these changes accelerate.
Bottom line, it’s important to get ahead of these changes as much as possible and prepare your current and future workforce to thrive in an environment where AI technologies are increasingly applied.
HR leaders should consider how to apply AI technologies to the products and services the HR function delivers to the organization. It’s easy to imagine this in the employee customer service delivery efforts of HR – it would be similar to how other parts of the business are looking at using AI technologies to enhance customer service for external customers.
HfS Research published an interesting paper with several examples of HR departments utilizing AI technologies in their product and service delivery: employees asking a virtual agent about changes in benefit coverage due to a new baby and the virtual agent responding with required steps and triggering downstream steps; parsing employee communications in various channels to identify dis-satisfaction and trigger suggested steps for the manager; increasingly sophisticated matching of employees and jobs with chat-bots managing much of the interaction with candidates. IBM’s head of talent acquisition presented its “AI-Augmented Talent Acquisition Journey” at a recent conference, demonstrating the significant investment they made in applying AI to this space. Companies like Volley are also applying AI technologies to create customized individual learning plans and games. Bernard Marr recently wrote a story in Forbes about a new product from PeopleDoc, the HR Service Delivery platform:
"[PeopleDoc] launched their solution to the problem automating the disparate toolsets used across business support functions. Their Robotic Process Automation platform uses ‘PeopleBots’ which run alongside existing systems to ‘listen’ for events and processes which can be automated. The idea is that they will use analytic technologies including machine learning algorithms to learn to automatically execute repetitive tasks and follow-on actions, leading to more accurate results and reducing the need for humans to carry out tedious repetitive actions."
Going back to the map of the technologies in the AI field of study, you can imagine many more ideas for changing and improving how HR delivers its various services to the company.
When I began this learning journey in early 2017, the business press had stories with a lot of hype about AI technologies, many provocative long-range estimates of impact – but limited examples that clarified how things would evolve. In just over a year, I now have more clarity about AI technologies and capabilities and have seen a proliferation of stories of very specific applications. The future is here now, and great HR leaders will be integrally involved in determining how to apply the technologies for the benefit of the customers, the employees and the company as a whole.
Interested in learning more about AI? Read Part One of our “De-Mystifying AI” series here.
About Todd Shaw:
Todd Shaw is a Human Resources (HR) executive and consultant who has a passion for building technology-driven service businesses that create valuable new opportunities for their customers. He has partnered with C-level leaders for over 20 years to help them build outstanding leadership teams, align culture with strategy, and drive change to achieve business plans. He has helped build organizations that delivered organic growth (PayPal revenue more than doubled during his tenure) and acquisition-driven growth (Bank of America revenues nearly tripled from 1996-2006). He has led global HR operations for over a dozen years, including 3.5 years living in Asia. To enable these results, he transformed multiple HR groups to enable the delivery of the required solutions and services. Todd has most often played broad HR leadership roles for businesses but also has focused at times in the areas of talent acquisition, organization & leadership development, and compensation.
Todd recently started his consultancy, Shaw Organization Design LLC, to focus on helping companies build high performing teams and organizations. Todd was most recently the SVP, Chief HR Officer at the global retail payments leader Verifone. At Verifone, he was focused on the transformation of the business from a hardware provider to a provider of services and solutions. Prior to Verifone, he served as a VP of HR at PayPal. At PayPal, he was focused on supporting the growth of a business doubling in revenue every three years while simultaneously transforming the technology organization to enable accelerated product development and delivery.
Prior to PayPal, Todd was an HR executive at Bank of America for several years. He worked in every major segment of their business, with his last assignment being the Regional HR Executive for Asia Pacific. He has also worked at the Taco Bell division of PepsiCo and at NCR. Todd holds a bachelor’s degree in finance and information systems from Drake University and an MBA from the University of Michigan.