Skip to main content Skip to footer
What’s new at monday.com

Start Small, Keep Control: monday.com’s Leaders on Adopting AI

monday.com• •5 min read
Start Small Keep Control mondaycoms Leaders on Adopting AI

Almost everyone at work uses AI now, but very few companies can point to what it has actually changed for the business.

Our own survey, conducted by Nielsen, found that while 72% of senior business leaders use AI at work, only 29% work at companies with a fully optimized and centralized AI infrastructure.

At Elevate ’26 in New York City, five monday.com leaders sat down with Rebecca Knight of theCUBE to talk about how to close that gap. Despite the differences in seniority and expertise, their advice is similar: start small, build on the work you already do, set limits early, and keep people in charge.

Start small, and start with your own team

Eran Zinman, co-founder and Co-CEO, on becoming our own customer zero

Watch the interview

Eran’s view is that the hard part of AI is the business change, not the technology. His challenge to leaders: imagine how you’d build your company if this technology had existed ten years ago, then use AI to improve what you already have or do something new you never thought possible.

monday.com took on that challenge first. In mid-2026, the company stopped regular work for a week so teams could build agents on our own platform. If the hard part is the business change, we wanted to experience it firsthand before asking customers to undergo the process. Teams throughout the company built more than 3,600 agents, and about 2,000 are still active. That experience serves as the foundation for monday.com’s entire approach to helping customers adopt AI.

Read more: Monday.com sees human-AI collaboration as the future of work

Casey George, Chief Revenue Officer, on seats + credits pricing

Watch the interview

Casey explained that customers know AI is valuable but don’t know where to start or how to get it deployed. That’s why monday.com is investing in services and forward deployed engineers, who will work directly with customers to identify an AI use-case and then build a proof of concept using the customer’s own data and systems.

The point of that approach is speed to a concrete result. Because the engineers build on the customer’s real data from day one, the customer sees something working almost immediately instead of running a long pilot. His closer: the limiting factor is now creativity, not technology.

Jillian Kamiyama, Director of Sales, on starting slow

Watch the interview

Jillian opens customer conversations with the problem, not the technology. A small window manufacturer in Georgia wanted to cut production times and see what was happening on the shop floor. An agent acting as a floor manager, flagging escalations and warning when supplies run low, turned out to be the answer. Customers are excited about getting time back, but what holds them back is cost control and, for larger enterprises, security and compliance.

Her advice is to be methodical. Trial AI with a few users and workstreams first, then whiteboard your repetitive workflows and ask, “If I could hire a human to do this, would I?”

Keeping control without slowing down

Ryan Nigro, Cybersecurity Trust and Enablement Team Lead, on governance as a speed booster

Watch the interview

Ryan’s point is that governance doesn’t have to be the team that says no. He compares AI governance processes to the brakes on a car: you can only drive fast when you trust you can stop. That’s why he recommends least-privilege access and rolling AI out piece by piece. When people know what can’t go wrong, they build faster.

Visibility is what makes that trust possible. Ryan points to an admin dashboard, activity logs and regular reviews as the way to keep a clear view of what agents are doing, including how AI vendors handle company data.

Read more: Brakes built for speed: Why least privilege lets monday.com’s AI teams move faster

Kate Nir, Product Director, AI workspace, on building an agent live

Watch the interview

Kate built a recruiting agent live on theCUBE, from one plain-language instruction. Ricky the recruiter agent reads the job brief, connects to the team’s hiring board to learn how the work runs, and recommends a next step, such as booking a phone screen for the top candidate. It was live the moment she clicked build.

With monday.com agents, control is built in. An agent states the permissions it needs, and owners can require approval before actions are taken, like sending an email. Any agent can be turned off, and every one of its actions is logged.

Kate’s advice for customers is to “start where it hurts:” start small, with one problem in your day that you wish someone could take off your plate.

The takeaway

Across these conversations, the advice is consistent. Start with a pain point, not with the technology. Keep a person accountable for the results, and know what it will cost before you scale.

That’s what monday.com is built to help with. If you don’t know where AI will pay off, we’ll help you find out. Our deployment experts work alongside your team to find the bottlenecks worth solving and get a first project running within weeks, if not earlier.

Want to see it in practice? Read how Lowe’s, Feed the Children, and Hewlett Packard Enterprise put AI to work. Or watch the opening keynote and read our updates from Elevate ’26. Sign up for updates ahead of Elevate ’27.

The interviews above were recorded by theCUBE, the video team at SiliconANGLE Media, and are sponsored by monday.com.

Get started