How software giant Workday got 79% of its employees to embrace AI

Leadership at business software giant Workday wanted employees to embrace artificial intelligence, but after conducting some internal research, they uncovered a few barriers.

Their study found that 43% of Workday’s employees—known as “Workmates”—said they lacked sufficient time to explore AI. More than a third of them also expressed uncertainty about how to use these new tools and worries about reliability and accuracy.

“Here we are wanting them so badly to explore, but they don’t feel that they have that time or that permission,” says Ashley Goldsmith, chief people officer at Workday. “What we’re working on is really changing the mindset.”

To encourage greater use across the organization, Workday held a splashy, all-hands meeting in April that prominently featured AI use case testimonials from across the workforce. Workday also set up a digital academy to promote AI upskilling and hosted a “prompt-a-thon” where employees could brainstorm problems they think can be solved with AI and develop prompts to best leverage large language models.

In another nudge this year, senior leadership for the first time mandated that all 19,300 employees establish personal goals for how they will use AI to improve their work and learn new skills. Their progress will be assessed by managers at the end of the year. 

Workday says these “Everyday AI” initiatives were built on internal analysis of the company’s workforce that uncovered that peer-to-peer guidance was more compelling than C-suite technologists evangelizing the benefits of AI. The company has also sought to reassure employees that experimentation is highly encouraged and that doing work faster with AI is always preferred over not using those tools.

“Everyday AI” was developed with the goal of boosting AI adoption across the company by 20% from the baseline set at the beginning of 2025. Workday says the increase was a better-than-anticipated 37% through May, with 79% of all workers now using AI. The tools used now range from the company’s own AI chatbot Workday Assistant to AI features from vendors including Zoom, Google, and Slack to generative AI-specific tools to support specific functions like customer support and coding assistant GitHub Copilot for developers.

Jim Stratton, who recently became Workday’s senior vice president of technology and architecture after serving as chief technology officer from 2018 until May this year, says his own approach to generative AI has evolved over the past few years.

Historically, the company would roll out fresh new features to all customers globally at the same time. But innovation is moving too quickly for AI—and some customers want to see early versions of AI-enabled tools before they are more broadly launched. That’s led to a staged rollout process for generative AI features, including at Workday, where early adopters get access to new tools first.

He’s focusing more on measuring the return on investments for generative AI, which can be easier to track for AI tools that assist customer support specialists or software developers using AI to generate code or bug fixes. But Stratton says ROI can be more difficult to quantify for other use cases, including when used to more accurately predict sales forecasts or when to help craft a pitch to a customer.

“Increasingly, in probably the last 18 months or so, there’s a real focus on measured ROI out of those investments,” Stratton says about AI and machine learning advancements. “Both in terms of what we do internally and also the products that we now go build.”

Workday says it has put extra emphasis on the company’s responsible AI principles, which include testing, risk assessments, and documentation, all work that’s especially critical for a software company whose tools are used to recruit and onboard talent, performance management, and onboarding. Some workplace tasks associated with this work, like decisions around compensation or promotions, should remain with workers. 

“There’s certain critical steps that for a very long time, I think humans will absolutely still be the decision makers,” says Stratton.

While that may be some comfort to human resource employees, fresh fears of AI’s impact on the workplace have increased in recent weeks, encapsulated by Anthropic CEO Dario Amodei’s warning that AI could eliminate around 50% of all entry-level, white-collar jobs. Workday itself generated headlines along those lines when it announced in February that it would lay off 1,750 workers, or 8.5% of its staff, as the company prioritized investments like AI. 

With developer productivity improving by 20% or more, Stratton acknowledges the fears workers may have that companies will need fewer employees to do the same amount of work. “That could be true,” he says. “But the way we view it, particularly on the development side of things, we can get more done with the same number of people so we can just go faster in terms of delivering more product.”

Goldsmith says there could be cases in which the technology completely takes over the work a person does, but ultimately she espouses AI’s benefits to both the business and workers. This is the tough sell that all businesses are confronting: encouraging workers to use AI to complete more tasks, while assuaging concerns that doing so won’t put them out of a job.

“We can reinvest those dollars in our technology and do more to advance the support and work for our customers,” says Goldsmith. “That’s how we talk to our employees about it. It is about super charging them, not replacing them.”

John Kell

Send thoughts or suggestions to CIO Intelligence here.

This story was originally featured on Fortune.com