Do investors and executives need to personally build with AI tools to judge the technology well?

Gustav Söderström · Oct 12, 2022

A technology leader must re-educate themselves deeply — math, code, scientific papers — when a fundamental technology shift arrives, rather than slipping into being just a manager

Without deep understanding you cannot credibly lead the change or evaluate what the company should do

Scope: costly in time for him and his family; examples given: deep learning around 2014-2015, and crypto

40:30 20 Product: Spotify's Gustav Söderström on Why Product is 100% Science and 0% Art, Why You Should Look at the Competition and then Do Something Completely Different & Why Talk is Cheap and Product Teams Should Do More of it; Structuring the Best Debate

Gustav Söderström · Oct 12, 2022

A leader should learn the technology themselves rather than only hiring the expertise in, because you feel like an imposter if you don't understand what you're responsible for

As an engineer at heart he finds it hard to respect a boss who has no clue what he's talking about, so he doesn't want to be that boss

Scope: framed as his personal disposition as an engineer

43:00 20 Product: Spotify's Gustav Söderström on Why Product is 100% Science and 0% Art, Why You Should Look at the Competition and then Do Something Completely Different & Why Talk is Cheap and Product Teams Should Do More of it; Structuring the Best Debate

Tomer Cohen · Dec 20, 2023

A product leader today must have the willingness and aptitude to go deep on AI, or they are not hireable into a product org

They need to understand the objective and how to build it, the non-deterministic nature of models and how to design knobs for a good experience, responsible data collection that fuels the product, and they need the velocity of learning to keep up

Scope: they need not be the engineer building the model; requirement is aptitude and willingness, not existing expertise

47:30 20VC Roundtable: Spotify, Adobe & Linkedin CPOs on How AI Changes The Future of Product, Why AI is Now the Product, How TikTok Changed Product, Why Cost is the Biggest Barrier to LLM Usage & Why Incumbents Can Adopt AI Faster Than Any Prior Innovation Cyc

Joelle Pineau · Nov 3, 2025

A leader must go deep and personally understand the technical work — having an amazing team is not a substitute

Watching Zuckerberg get into AI, the depth of his questions and his willingness not to coast informed everything he did afterwards

52:16 20VC: Cohere's Chief AI Officer on Why Scaling Laws Will Continue | Whether You Can Buy Success in AI with Talent Acquisitions | The Future of Synthetic Data & What It Means for Models | Why AI Coding is Akin to Image Generation in 2015 with Joelle Pineau

Harry Stebbings · Apr 14, 2026

Too many GPs are not doing the work, and the number who have never built anything with AI is astonishing

54:47 20VC: Anj Midha on Investing $300M into Anthropic | The Early Days of Anthropic & How 21 of 22 VCs Turned it Down | The Four Bottlenecks to Compute | What the China Has Smashed and Why We Should Be Worried

Harry Stebbings · Apr 14, 2026

If you are not building with AI tools yourself, you shouldn't be investing at all

You only learn where the real bottlenecks are — like integration friction — by building things yourself

54:54 20VC: Anj Midha on Investing $300M into Anthropic | The Early Days of Anthropic & How 21 of 22 VCs Turned it Down | The Four Bottlenecks to Compute | What the China Has Smashed and Why We Should Be Worried

Anjney Midha · Apr 14, 2026

With the right AI tools one person can now do what would have taken fifty people four years ago, and leaders who haven't personally built and deployed an agent don't understand what's happening

He runs a sovereign ministers' program and a Stanford class on the 'one person frontier lab' premise and won't certify participants until they build and deploy agents

55:17 Must ship an agent yourself to grasp the shift