GM Financial’s business case for making AI fun for employees

Talking about AI can often be a somber endeavor, with lots of conversation about tech stacks and ethics and ROI.

But it doesn’t have to be quite so serious.

With the right attitude, implementing AI can be a fun process that employees will want to get involved in.

GM Financial, a regulated company that employees call “risk averse,” found a way to make implementing and interacting with AI an enjoyable, playful experience.

For instance, they don’t talk about AI as some abstract chatbot. They talk about Coco.

“My entire team talks about Coco like she is a human, like she is a full team member,” explained Carlye Greene, vice president, IT Technical Learning and Readiness at GM Financial, said during a panel at Ragan’s AI Horizons Conference Tuesday. “I was checking email before this, and I got a note from one of my team members, and he was like, ‘Well, Coco took a first draft of this.’ We just kind of make it fun.”

That fun was by design.

By humanizing their AI tool, the cutting-edge technology becomes approachable rather than standoffish and scary. That’s just one part of GM Financial’s commitment to making employee adoption of AI easy and entertaining.

Balancing play with practicality

When implementing artificial intelligence, some organizations default to telling employees to “just go play” with the tools. This isn’t a bad strategy per se, but in a regulated industry, it can come with risks: what people develop may not wind up being useful if it isn’t developed within useful parameters.

GM Financial ensured that experimentation time was productive by hosting a hackathon with clear guidelines based on both tech safety and business cases.

An AI subcommittee identified automationuse cases that could have the most business impact and the most utility based on the approved tools. From there, workers from both business and IT were brought together to solve these challenges. IT brought the tech know-how while business brought the customer knowledge.

The best solutions got to present to the CEO and eventually went into production.

“Often, I think we say, ‘Hey, go play with this,’ but then nothing comes of it,” explained Greene. “There’s no business value there. So we were able to bring all of those pieces together in this companywide, sponsored hackathon.”

The freedom within constraints

Another aspect of keeping AI learning lightweight was a conscious choice to limit the AI tech stack to just a few tools. GM Financial standardized on an Azure-based AI stack, and focused on teaching the core competencies workers needed to succeed.

“We do a curated, skills-based learning approach, which basically means adults are really busy,” Greene said. “They hate being told what to learn, and then they hate learning things that they aren’t going to use.”

To focus on the most important training, Greene and her team partnered closely with IT to understand their technology roadmap to curate training tied to that plan.

“We’re not going to teach an AI tool that’s not available. We’re going to teach the AI tool that’s available to the person, hopefully right as they’re getting access to that tool or while they’re getting onboarded into that tool,” Greene said.

This strict focus allows workers to focus on homing their ability in a few key areas rather than becoming generalists. The partnership with IT ensures each training is tailored to what workers need, not nice-to-know specifics that will fade in memories with disuse.

That focus didn’t happen by accident. It was part of a broader effort to make AI approachable from day one.

“You can’t have a tech transformation without a talent transformation,” Greene said. “The two have to go hand in hand.”

For more on these topics, visit the Center for AI Strategy. 

Allison Carter is editorial director of PR Daily and Ragan.com. Follow her on LinkedIn.