How AI is Reshaping Product Operations
Lessons from the AI Product Summit
AI is transforming product work by taking on tasks that slow teams down, from synthesizing feedback and surfacing risks to generating early prototypes that help product managers (PMs) move with more precision. This shift has made product operations the group responsible for integrating AI into the systems PMs already depend on.
Productboard’s AI Product Summit brought that message into focus through four perspectives that landed on the same truth: AI is reshaping how product decisions get made, how work moves through an organization, and how product ops can guide teams toward real value.
Everyone is trying to understand where AI fits in the product lifecycle. What emerged across the talks was a set of pillars that explain how AI can accelerate product work and where product ops can lead the transformation.
Pillar 1: AI Is the New Collaborator for Product Decisions
Chris Butler, Director of Product Operations at GitHub, broke down the decision-making lifecycle—from identification to communication to learning—and showed how AI can enhance each step, without replacing human judgment.
He emphasized that product ops professionals must shift their thinking from static documentation to living, AI-enhanced decision systems. At GitHub, his team is already building AI agents that triage intake requests, critique product specs, simulate missing stakeholder perspectives, and even flag violated assumptions in real time. These agents actively shape product conversations.
Chris urged teams to be intentional about how they decide. That means choosing when consensus works, when vetoes are faster, and where automation can spot patterns humans miss.
“Deciding how to decide is actually one of the most important things you can do for a decision-making process.”
Chris Butler
Director of Product Operations, GitHub
He also offered a practical, and interesting, tip: build “mean” prompts. Inside GitHub, they use sarcastic, bot-delivered critiques to challenge ideas without hurting feelings. Because it’s easier to accept harsh feedback from a machine than a teammate.
For product ops, the call to action is clear: treat AI as a partner in process design, not just a speed boost. Start by mapping how decisions get made today. Then look for the handoffs, bottlenecks, and blind spots where AI can add value.
Pillar 2: AI Is Changing PM Skills and Expectations
GenAI is changing the definition of a “good” product manager. Craig Mirsky, Senior Director of Product Management at S&P Global Market Intelligence, shared how product ops can help PMs evolve by focusing on three imperatives:Â
- Integration: Consider how AI impacts all aspects of work. View AI not just as something delivered to end users, but as something to empower the team.
- Ecosystem thinking: Identify and break down silos to improve client workflows within current constraints.
- Continuous training: Constantly reinforce and learn. GenAI is an iterative process.
At S&P, Craig’s team supports hundreds of enterprise PMs across siloed teams and complex data products. GenAI, he said, breaks those silos. It requires PMs to think in terms of connected workflows—not discrete features—and pushes product ops to act as both enablers and educators.
That means helping PMs develop practical skills like prompt engineering, AI-driven research, and technical communication. Change should come from the top. Craig said that once their Chief Product Officer was bought in on the importance of learning these tech skills, everything accelerated.Â
To help PMs keep pace, Craig’s team sends biweekly AI-curated newsletters, builds internal workshops, and runs hands-on experimentation sessions. He called GenAI a “second brain” for PMs, but warned that without real curiosity, no tool will help.
“We are engaging our team members to go learn—bring examples forward, show us how they’re thinking about design and AI… If you don't have a curious product manager, you'll struggle to get the product to market.”
Craig Mirsky
Senior Director of Product Management, S&P Global Markets
Build the scaffolding, not just the strategy. Help PMs learn by doing and monitor those skill gaps. It’s all about pushing your org to treat AI like a core competency, not a side quest.
Pillar 3: AI Workflows Give Product Ops New Powers
Ross Webb, Founder of Product Team Success, kicked off his session with a startling statistic: while 87% of product ops teams report AI initiatives in progress, only 12% are seeing real impact. AI is trapped in silos. With loads of single-point solutions getting introduced into everyone’s tech stack, Ross made a point that you’re “multiplying compute costs across the org without delivering actual outcomes.”
To fix this, product ops needs to stop thinking in tools and start thinking in systems. Ross introduced a three-level AI maturity model: reactive, systematic, and strategic. According to him, most teams get stuck in reactive mode. They are struggling to prove ROI because they’re running disconnected pilots and dabbling with AI features.Â
The shift from reactive to strategic happens when product ops steps up as a force multiplier, building agentic AI systems that span the entire stack.
“You need to work out how to get that five to 10 hours saved per product manager per week [to prove ROI]”
Ross Webb
Founder, Product Team Success
That means orchestrating agents that talk to each other. Not just summarizing a PRD or writing a user story, but connecting roadmaps, OKRs, feedback tools, analytics platforms, and dev boards into a cohesive “AI brain.” In one real-world example, Ross’s team connected Productboard, Linear, and PostHog to surface a key product funnel drop-off and auto-generate a prioritized roadmap with proposed actions. It took about five minutes—not weeks—to uncover.
The message for product ops? Don’t wait for your PMs to ask for help. Build the infrastructure now. Start small. Audit your stack. Pick one high-friction process and create a focused agent. Measure everything. Then scale.
Accelerate Your Product Processes
Whether shaping smarter decisions, upskilling PMs, connecting fragmented workflows, or rearchitecting how teams operate, product ops is where AI becomes real.
But AI won’t transform product work through experimentation alone. It takes intentional design and a team willing to pilot, fail, learn, and scale.
You don’t need to overhaul everything at once. Start with one problem. Build one agent. Pilot one prompt. Then measure what matters.
Watch the full AI Product Summit
If you want the full depth of insight from these product leaders, watch the recorded sessions from the AI Product Summit. Their talks show where product work is heading and how product ops can help teams move with confidence.
