How Walmart is securing a new cohort of AI builders

It’s all fun and games until the majority of your workforce suddenly becomes AI builders—and needs new and improved guardrails to prevent security havoc.

That’s the predicament faced by numerous companies thanks to the rise of vibe coding, and retail giant Walmart is no exception.

All in on AI. Walmart has been gearing up its workforce for the AI era. During a June 3 press briefing at its annual Associates Week, Walmart Global Chief Technology Officer and Chief Development Officer Suresh Kumar said associates are now leveraging AI tools like Code Puppy, Walmart’s internal vibe-coding platform.

“Right now, we have as many non-technical associates who are now able to do things that require technology as we have engineers,” Kumar said. The retailer also announced the general availability of its OpenAI certification program to US associates to bolster AI literacy across its workforce.

“This new certification helps associates build practical competence with AI: how to use AI tools effectively, how to evaluate the output that those tools produce, how to understand the limitations, and how to keep human judgment at the center,” Josh Allen, Walmart group director of learning strategy, said during a June 4 press gathering at Associates Week.

Securing a workforce of builders. With new builders, however, comes new problems. Akshay Bhargava, VP of product management for AI software and platform at Cisco, told IT Brew the “democratization” of AI is changing the game for security professionals.

“This concept that everyone’s a builder, it does break the traditional security models,” Bhargava said in an April interview, noting a growing shift away from a standard software development life cycle. The uptick in citizen developers also paves the way for a rise in shadow AI, he added.

“Now with builders, they can be creating thousands of agents. I’m talking to engineers [who] have thousands of agents that are working simultaneously,” Bhargava said. “So, now you can imagine that the order of magnitude of shadow AI is exponentially growing.”

Walmart’s approach. The task of securing a new generation of non-technical developers is top of mind for Kumar, who told IT Brew in an interview that “the best way to drive governance and security is to make sure that it’s built” into the tools themselves.

One way Walmart does this, he added, is with the help of Element, an AI platform that the retailer’s developers use to build and test AI applications.

“Element allows us to plug multiple different LLMs in, and Code Puppy will talk to an LLM that has already been vetted, that has already gone through our normal security review,” Kumar said.

Kumar added Walmart uses a shared internal platform that allows it to identify points of friction associates may run into while using Code Puppy, along with the ability to redirect them to more efficient tools.

“We can see if people are doing the same thing again and again…[and] we can build things in so that they can do it a lot more efficiently,” he said.

Granting Code Puppy users with appropriate permissions and access is also crucial, Kumar added.

“Just because you are an associate in Walmart, you don’t get access to everything,” Kumar said. “Code Puppy understands that, and Code Puppy will give you access to the information that you would have normally had access to anyway.”