Which skill matters most for PMs as AI speeds development?

Sam Stephenson

Founder · Granola

Knowing what to build

Sam Stephenson says the hard part is knowing what to make, since making things is now so easy; that thinking needs space, not constant grinding.

Elias Lieberich

Founder & Product Leadership Coach · Product Matters

Judgment

Elias Lieberich says judgment is the key skill, since AI tools help build and analyze, but human judgment decides which of the 50 prototypes to move forward.

Ravi Mehta

Angel Investor · Traveljoy

Judgment

Ravi Mehta says judgment — alongside craft and empathy for the person on the other side of the screen — because AI didn't take the PM job, it exposed the work PMs should have been doing all along.

Geoff Charles

CPO · Ramp

Taste making

Geoff Charles says the taste maker role is arguably among the most important for PMs, holding the steering wheel and the bar for great taste, doing what AI cannot.

JE

Jonathan Evens

Product Lead · Google Deepmind

User empathy

Jonathan Evens says user empathy, focusing on deep user problems and starting with the problem, is essential for AI product or feature PMs.

Ben Celebicic

Chief Product and Technology Officer (CPTO) · Hinge

Rapid prototyping

Ben Celebicic says rapid prototyping is the big gain: without engineering support he can build something close to the final product, put it in users' hands, and quickly learn whether it's worth building.

Lars Grønnegaard Hansen

CPO and Co-Founder · Dreamdata

Listening to sales calls

Lars Grønnegaard Hansen says listening to sales calls is the practice he should have mentioned first, and AI makes it scalable—PMs can mine transcribed calls to learn what customers ask for.

The answer starts at 38:07.

Tara Seshan

Member of Product Staff · Openai

Elevating ambition

Tara Seshan says that part of a PM's job now is to elevate everyone's ambitions, asking whether the possibility ceiling is meaningfully higher or the work could be done faster, which she calls a huge part of the product management role.

Jeff Chow

Chief Product and Technology Officer · Miro

Communication

Jeff Chow says the skill of a product leader is communicating, and that product teams should use prototyping as a tool, not a tax, and use it in the right way.

Sachin Rekhi

Course Creator · Reforge

Market research

Sachin Rekhi points to market research, saying AI is good at it and he built skills around it, such as a weekly competitor pricing report.

Hilary Gridley

Founder

Using AI daily

Hilary Gridley says the most important thing is to use AI a little bit every single day, because there is an activation-energy hump of just sitting down and doing it.

Srinivasan Raghavan

Chief Product Officer · Freshworks

Customer focus

Srinivasan Raghavan answers customer focus, saying AI lets builders understand what users need and what customers would love to use, and his team synthesizes customer conversations to decide what to build.

Oji Udezue

Author · Productmind

Judgment

Oji Udezue says the scarce skill is judgment—saying this one and not that one, which he calls Taste at Speed—because PMs cannot prompt or hire an agent to hold it and must own it.

Mina Doroudi

Principal Product Manager · Microsoft

Innovation

Mina Doroudi says that if only one of innovation, design and passion matters, it is mainly innovation, because she is trying to put candidates on the bleeding edge and see how they react into unknown territories.

TO

Todd Olson

CEO / Founder · Pendo.Io

Experimentation

Todd Olson says experimentation is one of the big soft skills, because people who try many things are in a heyday and teams still need a lot of experimentation.

FF

Faith Forster

Chief Maker · Discoveree For Product Leaders & Managers

Working alongside customers

Faith Forster says PMs should work alongside customers, watching how they use the product, because AI makes shipping easier and shifts the definition of done to adoption.

Dianne Na Penn

Head of Product Management, Research · Anthropic

Hands-on building

Dianne Na Penn says product leaders must be hands-on themselves, spending time actually shipping with the technology and sweating the details alongside their teams, because people can't see what a good AI product looks like unless they've built one themselves.