We have all heard about the potential for Generative AI to unlock productivity gains not seen since the Industrial Revolution. But adoption comes before productivity, and it takes more than a bunch of enterprise licenses for a workforce transformation. 

In a recent webinar, Diane Gherson, former CHRO of IBM, and Jennifer Carpenter, Global Head of Talent at Analog Devices emphasized that for HR leaders, this means we need to help our people adopt and adapt to AI. In 2011, way before the public launch of today’s biggest LLMs, Diane rolled out an AI HRBP at IBM that went on to receive 1.65 million queries per year. At Analog Devices, Jennifer has been an early adopter of generative AI, and for the past three years, has used it to streamline and improve performance reviews. She is now also in the process of scaling Valence’s AI coach, Nadia, across the organization with best-in-class adoption and engagement metrics. 

Here are 7 of their most important takeaways for any HR leader who wants to drive AI adoption, transform how work happens, and impact the bottom line. ‍

1. Design the work with your people

At IBM, when Diane rolled out her HR business partner, the entire workflow changed, and she focused on redesigning the work with her people and helping them upskill for the new jobs to be done.

“One of the things that we discovered at IBM was that using an agile methodology actually engaged our people in the design of the work. And after all, you can't redesign work without the people. They are the ones who actually know the work, and they understand the workflow.

What we found was so important was to be able to say, our attrition rate is 14% in payroll. We won't replace those people, but your jobs are gonna be okay. You need to upskill to get the new jobs that we'll be creating because AI is going to take over some of the more sort of menial tasks that you're all doing, so that you can then be looking at patterns and broader issues and helping us actually improve how we do compensation and benefits in the process.”

 - Diane Gherson

2. Let people opt in to your AI pilots

Establishing trust starts with how you invite people into the AI pilot process. As Jennifer learned early on, no one is going to throw you a parade just for giving them an LLM license.

“We shifted gears to an invite-only approach, and we've seen great success with that model because when people say, ‘I'll try this out,’ they're more likely to actually go to the training and feel a sense of accountability to tell you what they think because they accepted your invitation.”

‍ - Jennifer Carpenter

3. Learn what’s behind the choice to opt in or out

Jennifer has surveyed her employees on AI with simple questions to gauge their optimism and confidence and understand why they do or don’t opt-in. She now has a dataset of 2,000 responses to inform her AI strategy going forward.

 
“We ask the same questions to those that opted in and opted out. We asked employees to what degree do you agree with the following statements? I believe generative AI is going to improve my productivity. I believe generative AI is going to improve the quality of my work. I'm confident that I know how to use these tools, that I can navigate them effectively.

And what we saw is those that did accept your invitation were statistically more optimistic and more confident than those that did not. And we also asked them, why are you opting out? And I was really struck that 44% said I just don't have time. So that's a barrier, this perception that I just don't have time to even try it. And then the other categories: 14% said ‘I would prefer a human coach.’ 14% said ‘I don't believe this is a great use case for AI.’ And 14% said ‘I tried AI, I had a bad experience, and I'm not interested in trying again.’”

- Jennifer Carpenter

4. Build trust by modeling the change

Leaders, if you want your employees to make AI integration a priority, you have to show them that it’s a priority for you, too. That means showing up to trainings just like everyone else.

“If someone’s manager is saying, ‘Oh my god, that third module was a killer, but I got through it,’ that person's thinking, ‘Okay, well, my manager found the time and they got frustrated, so maybe I could get through this too.’”

- Diane Gherson

5. Use what you learn from early adopters to move from individual use to collective adoption, where the real productivity gains happen

“The early phase of Gen AI adoption was at a very individual level: It's good for me. It's gonna make me more effective. But you can't really unlock the productivity improvement without actually engaging a whole team and doing work differently. And that's a very different thought. I think that's the piece that we need to focus more on than the individual piece. 

Think about three different audiences, and work to improve the lives of those three. The first one is the manager. Managers are overwhelmed. There's so many ways that we can help managers with AI. The second group is the employee. How do we make their work lives better? And the third is the business. Are we actually moving the needle on productivity and quality? Be sure to be measuring those things.”

- Diane Gherson

6. When looking for initial use cases, find work that is a drag on the organization

When work feels heavy, people tend to avoid it. But when that work suddenly gets lighter, people put more effort into it, boosting both efficiency and quality. Annual performance reviews are a great example, and one of many use cases that Analog Devices has given employees the option to use AI.

“Over 9,500 employees engaged. 94% said it improved the quality of their reviews, and 76% said they got the work done in half the time.”

- Jennifer Carpenter

7. Be willing to experiment and be vulnerable

As each week brings new releases and features, the difference between just paving the cow path and a powerful transformation comes down to running a lot of experiments with a lot of different tools. Not every one of these is going to succeed, and that’s okay.

“At the heart of experimentation is: you win some and you lose some. But you have to keep trying. You don't always get it right, especially with adoption, because we're at the beginning of this journey. With rapid experimentation, leaders have to be vulnerable. Perfect should not be our goal. It should be practice.”

- Jennifer Carpenter

[[button-cta]]
[[divider]]

Jennifer Carpenter is Vice President and Global Head of Talent at Analog Devices. She previously served as Vice President of Talent Acquisition at IBM and Delta Air Lines and spent over two decades leading HR, Learning and Talent Development, and Talent Acquisition at Accenture.

Diane Gherson is the former CHRO of IBM and now serves as a Senior Advisor at BCG, as Board member of The Kraft Heinz Company, Centivo, Techwolf, National Academy of Human Resources, and as a member of the Executive Advisory Board of Semper Virens Venture Capital, a VC firm specializing in early-stage human capital management startups.

Parker Mitchell is the Founder and CEO of Valence.