Disruptive technologies such as artificial intelligence, blockchain and augmented/virtual reality have the potential to revolutionize not only HR but the nature of work itself.
Tim Gregory, the director of HR innovation and workforce technology at Corning Inc., says HR organizations need to develop a thoughtful approach to evaluate disruptive technology and its potential impact. He would know — during his career, Gregory has led efforts to digitally transform HR for three Fortune 500 chief HR officers, including two National Academy of Human Resources fellows.
Two years ago he began focused efforts to identify the practical application of AI technologies within HR. He led co-development efforts with IBM to prototype HR solutions powered by IBM’s Watson.
He will be speaking at UNLEASH America in Las Vegas on May 15-16. We spoke with him recently about his upcoming talk, his work at Corning and the dynamic technology landscape that surrounds HR.
What are you going to be talking about at UNLEASH America in May?
We’re going to focus on two topics. We’ll start by sharing key learnings from our global-scale HR Digital Transformation project, in which we essentially leveraged the entire suite of SAP SuccessFactors products across 22 countries, in 12 languages with over 70 integrations. We’re going to talk about the top three things we did right and the top three things we did wrong.
That’s the first part. Then we’re going to talk about what’s on the horizon — what does the view look like after you have made it to the cloud? We’ve been exploring a variety of disruptive technologies, including artificial intelligence, data lakes and robotic process automation. For UNLEASH America in May we’re going to focus on something I think people are really interested in that will help them cut through the vast amounts of hype that is surrounding artificial intelligence today. We will demystify AI and provide the audience with a practical framework that they can use to assess whether or not the technology products that they are evaluating are indeed AI-enabled.
What is your focus these days, and what technologies are you exploring in your current position?
When I joined Corning back in 2014, we developed a strategy that we called the “HR2020 Roadmap.” In it we defined a number of principles, attributes and measures that we wanted to achieve prior to the year 2020. To achieve these business goals, we recognized that we had to adopt a modern, agile and cloud-based technology platform. We chose SAP SuccessFactors and it has delivered a lot of value for us. The mobile app delivers frictionless direct access for employees and managers, which, in turn, enables higher levels of engagement. We also benefit from the quarterly release of new features and capabilities that are included in our annual subscription.
My focus today is on developing our “HR2025 Roadmap” which evaluates the practical use of the disruptive technology I mentioned earlier as well as augmented reality, which has enormous potential in the learning space, and blockchain, which has the potential to fundamentally change how the world accesses and manages personal data.
How do HR leaders sort through all of this new technology to find tools that actually add value?
I have benefited a great deal from the world-class instruction that is broadly available online from institutions like MIT and Cornell. I’m also a strong believer in the idea that no one can genuinely understand something new until they can use it. So I think it is really important for HR leaders to commit to lifelong learning and intellectually engage in efforts to develop practical HR use-cases and prototypes.
I also rely a great deal on my network, which I actively invest and nurture by attending conferences, taking the time to meet with new and established vendors, and developing and growing my social media connections via LinkedIn and Twitter. Social media has gotten a bad rap lately, but when used intelligently it can be a phenomenal learning platform capable of enabling amazing access to some of the world’s leading subject-matter experts. It takes time and effort going through the articles and posts, seeing who the authors are, who they’re referencing, and then finding them online and seeing who they’re following to triangulate where these innovations are coming from. But once you’ve tuned your social network in to those channels it’s phenomenally valuable and I rely on it a great deal.
With a little effort and time you develop a much keener sense for what’s pixie dust and what’s genuinely practical. … But no matter how you cut it, when it comes to understanding disruptive technology and challenging your “comfort zone” you must grow your network and connect to people who are new to you and outside of your traditional organizational boundaries.
Should people in HR be more anxious or optimistic about the integration of AI into their field?
I think if people really understood what it was, they’d be very optimistic — and I think you’ll have those who lean forward and embrace the technology and they will do very well as a result.
If you go back to the early 2000s, people were hesitant about adopting the internet. There was a time when we were told “Don’t meet strangers on the internet and don’t give them your credit card information.” Today we routinely do precisely that and then we get in a car with them — we just call it Uber. The internet changed everything and genuinely transformed our society. Some of the emerging technology we’ve been talking about has the potential to drive society-level changes.
AI is one of those technologies. If you peer inside the core of today’s AI, you won’t find the Terminator. What you will find is vast quantities of data, statistical algorithms and cheap computing power. What is important to understand is that even the most advanced AI systems are not actually “thinking” at all. … No researcher or scientist who understands this well would describe this technology as being anywhere near human intellect. These systems don’t think; they compute, and they do this for the purposes of generating predictions. Which can be very helpful to us. Typically the human brain doesn’t do well with processing exponentially large numbers because these are not the types of numbers we encounter in our daily experience. AI-enabled systems, on the other hand, do that exceptionally well. HR organizations that learn how to build “human-in-the-loop” processes that blend the best capabilities of AI and humans will gain a distinct advantage in the war for talent.
For example, in today’s economy if you want to find the best talent you need to search everywhere, and you need take advantage of the vast quantities of useful data that potential candidates have readily made available on the internet. In this circumstance if you take a recruiter equipped with a trainable AI-enabled system versus a recruiter armed only with a browser and keyword searches, the AI-enabled recruiter will build the better candidate slate every time and will do so in a fraction of the time.
Despite all the promise AI-enabled systems have, they should not operate autonomously. The value of AI-enabled systems is limited by the quality of the underlying data and algorithms used to generate them. If there is bias in the data or the algorithms then the resulting prediction will be biased. For this reason alone, companies need to be very cautious about taking the human out of human resources and subscribe to “human-in-the-loop” models.
How willing have companies been to adopt these emerging technologies to date?
I think we’re at the very beginning of this. There are some leading HR organizations who are exploring it, but I think there is a lot of work ahead of us before it becomes “normal” to see AI used to its full potential, broadly across HR. We still have some key technical and “people”-related hurdles to overcome.
It’s probably not helpful to get too deep into the technical hurdles in this interview but I think it is valuable to at least recognize that AI is based on data and it suffers from many of the same weaknesses that traditional data analytics suffers from when the necessary data is sparse and of poor quality. In terms of “people”-related hurdles I think there are some very big questions that the industry needs to be able to solve for, like the potential for algorithmic bias as we mentioned earlier.
Another big problem is known as “explainability.” With the more advanced neural-net-based machine-learning models it is very difficult for even the data scientists, who developed them, to explain why a certain prediction was made. This is not acceptable in the HR space. The source of evidence used by the machine to offer a prediction needs to be auditable and, if necessary, the weighting of its contribution to the prediction easily changed.
Until these and other related topics are understood by a much larger percentage of HR practitioners I think it’s going to slow the broad adoption of AI. But there are a lot of very bright people actively working on solving these problems, and I think once they’re resolved you’ll see adoption on a massive scale.
The views, thoughts and opinions expressed in the text belong solely to Gregory and not necessarily to his employer, organization, committee or other group or individual.