For big companies seeking ROI from AI, people matter more than models
Fortune Jeremy Kahn ● Covered by 2 sources
Big companies say AI wins come from people, not just models. The real gap is training, hiring and letting employees actually use the tools.
Based on reporting by Fortune, Jeremy Kahn — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
At Fortune’s AIQ Summit in New York, one message kept coming back: big-company AI success is less about the model and more about the humans around it. The event, held at the New York Stock Exchange, focused on how Fortune 500 firms are getting AI working at scale, and the best examples were about training, hiring and organizational habits, not shiny demos.
ServiceNow’s Diana David pointed to a stark split in the companies Fortune classifies as AI “Pacesetters” and everyone else. Fifty-seven percent of Pacesetters are investing in upskilling employees on AI, versus just 4% of the rest. The same pattern shows up in talent plans: 68% of Pacesetters are working to attract, hire and retain AI talent, while only 10% of other companies are doing so. Pacesetters also stand out for long-term HR strategies built around AI and for assessing AI skills across the organization.
Amy Webb said a lot of enterprise AI frustration comes from companies not training people well enough to use the technology. She argued that large organizations get stuck in “learned helplessness” around tech. Drew Holler, the chief human resources officer at Lennar, agreed that companies have to provide tools and training, but he also said employees have to do their part and upskill themselves.
The theme from several speakers was adaptability over pedigree. Webb said the key isn’t being AI native but being flexible. Danielle Gonzalez, Palo Alto Networks’ chief people officer, said the company looks for people with agency who can learn quickly and unlearn what they thought was true. Palo Alto Networks now uses observable interviews and hackathons to see how candidates solve problems in practice.
There was also a clear skepticism toward the idea that governance concerns should choke off deployment. Webb said chief risk officers and CISOs need a seat at the table, but they should stop saying no before they say tell me more. That tension matters because the summit’s mood was practical: companies want returns, and the room seemed more interested in implementation than in endless caution.
Elsewhere, the supposed end of traditional SaaS looked overstated. Runway’s Michelle Kwon said the startup still buys off-the-shelf software for things like payments and HR compliance, even while it uses AI heavily inside the company. The broader message was simple: AI may change what gets built, but it doesn’t mean every business suddenly wants to reinvent payroll.
My take — AI-written commentary, not fact-checked reporting
The loudest AI skeptics in big companies are often the people whose job is to keep everyone safely parked in neutral. That may be comforting, but it is also how you get a very expensive slide deck and not much else. The smarter play is boring: train people, hire for adaptability, and let the risk team stop acting like the department of no.
Read more about this at: Fortune
