← All Episodes
Based on Lenny's Podcast data
Lenny's Knowledge Sketch

AI Will Transform
Your Product Forever

Paul Adams
Chief Product Officer, Intercom
FEB 2024
The Stakes

This Is a Meteor

TECHNOLOGY SHIFT
"This is a meteor coming towards you. This is going to radically transform society. And I think if people don't explore AI properly, it will leave them behind."
  • AI is bigger than mobile as a technology shift — maybe as big as the internet itself
  • This is not hype like crypto or Web3; it's a fundamental reshaping of how products work
  • If you don't engage with it, you will be left behind by the market
  • ChatGPT changed everything on November 29, 2023 — nothing feels the same after
Framework

Start with Your Problem, Then Ask: Can AI Do It?

CORE PROBLEMCAN AI DO IT?THREE PATHSYES → ReplacePartial → AugmentNOT YET → Wait
  • Step 1: "What's the core premise? Why do people use your product? What problem does it solve?"
  • Step 2: Ask, "Can AI do that?"
  • Step 3: Match your product to AI's actual capability
Paul's core framework

Replacement: AI just does it. Your old workflow is gone.

Or augmentation:

Augmentation: AI helps humans do it better, faster, with more confidence.

Timeline is key

Some use cases aren't ready yet. That doesn't mean ignore them — but know where you are on the curve.

Lessons from Failure

What the Froze-on-Stage Story Taught Me

  • Cannes keynote disaster: Froze mid-talk in front of thousands, cursed on mic, walked off-stage
  • Google+ failure: Built from competitive fear, not customer need. Privacy disaster.
  • Google Buzz failure: Same root cause — fear-driven, not user-driven
  • The pattern: Every failure traced back to motivations. Fear-driven building = failure. User-driven = success
The recovery

You can freeze on the world's biggest stage, panic, walk off, and still survive. Then you keep going. That perspective shift is everything.

Ship to Learn

Intercom's philosophy: Ship fast, ship early, ship often. Failure is built into the model. You learn faster if you fail faster.

Playbook

How to Go All-In on AI

  • Read constantly. Twitter, newsletters (Matt Rickard), OpenAI blog, papers. No excuse.
  • Play with the tools. ChatGPT, Bard, GPT Vision, Rewind. Don't just read — do.
  • Don't bolt it on. Don't create a separate "AI team." Integrate AI thinking into every product team.
  • Stay generalist. Specialists matter for ML, but generalists who learn AI faster win.
The time investmentBook half an hour daily to read, play, experiment. No crazy hours needed. Just discipline and prioritization. If you don't, it will pass you by.
Contrarian

AI Product Myths That Will Kill You

"It's just hype like crypto"INSTEAD →AI will reshape your product category. Reporting tools, project management, support software — all vulnerable to AI-native redesigns.
"We'll add an AI team to handle this"INSTEAD →Every product team must understand AI. Siloing it kills you. This is like mobile in 2010 — everyone needs to own it.
"We don't have time to explore AI"INSTEAD →You don't have time NOT to. 30 minutes daily. Read, play, experiment. Or you're already behind.
"We're not sure if this path is right"INSTEAD →No one knows. This is product taste + experience + educated guessing. Move anyway. Adapt or die.
𝕏︎ X / Twitterin LinkedIn📸 Instagram🔗 Copy link
0:00