Intel / podcast

The AI paradox: More automation, more humans, more work | Dan Shipper

Source: lennysnewsletter.com
Author: Lenny Rachitsky
Date: Published May 24, 2026

Context

An automated can create an ongoing maintenance job for a person. Dan Shipper, cofounder of the media and software company Every, says his team's experiments changed his preference from personal agents for everyone toward a shared company agent with a clear human owner.

His explanation is operational: someone must watch whether the agent remains useful, maintain its and fix failures. When that work becomes burdensome, people abandon their individual setups. A shared agent concentrates maintenance responsibility while serving more colleagues. This is Shipper's account of what works for his team and informs his forecast; it does not establish how automation will affect employment across the economy.

Terms in this piece · Glossary

AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
context window — The maximum amount of text a model can consider at once — its working memory for the current conversation or task.

Key quotes

“I'm simultaneously extremely AI-pilled and very bullish on humans.”
Dan Shipper

“What models do in general is they make yesterday's human competence cheap.”
Dan Shipper

“What agents do is increase the number of users of SaaS, not get rid of it.”
Dan Shipper

“We speed ran the CLI era. It was nice while it lasted, but I think CLIs are over.”
Dan Shipper

“And the mechanism is agents need people who care about them.”
Dan Shipper

Chapters

00:00Introduction to Dan Shipper
02:56Dan's unique position living in the AI future
09:17How the way we work will change in the coming year
16:39The case for general agents
18:08Codex and Claude Code as the new operating system for work
25:39How Cursor fits in
27:42How this changes what SaaS companies should build
31:13Why CLI is already over
33:34Two agents are better than one
36:22Why Dan is bullish on SaaS stocks
39:01Why automation doesn't reduce human work
47:00The value of human-written code
48:36Quick recap
50:15How work is changing
56:17Why data scientists are drowning in bad analysis
58:24Which product/tech roles are least changed by AI
1:02:17We will read way more AI-generated writing
1:08:28Why product managers will dominate the AI era
1:11:05Full-stack designers are the other big winners
1:13:11The AI job apocalypse won't happen
1:16:00How to ride the models to stay relevant
1:21:02Final predictions and advice
1:25:24Lightning round

Read the source www.lennysnewsletter.com