Oct 19, 2026 · 1h 3m
AI agents move from assistants to economic actors
What Happens When Al Agents Can Spend Your Money?
The episode examines how trust, integration, and product design could determine whether AI agents merely organize information or actively shape purchases, benefits, and marketplaces.
- 1Successful AI products may depend more on workflow integration, education, and trust than on model intelligence alone.
- 2Town argues that consumer-grade experiences can drive enterprise adoption more effectively than compliance-first software design.
- 3Agents that act on behalf of consumers could reduce the value of human attention and reshape marketplace distribution.
Don't miss
Rampell describes how Town recovered a missed dinner invitation buried in email, illustrating the value of an assistant with broad personal context.
The brief
Jean-Denis Greze moved from Plaid and an abandoned SMB tax product to Town, an AI assistant built around email, calendars, and recurring knowledge-work tasks.
Greze and Alex Rampell argue that AI adoption depends less on raw model capability than on integration, education, trust, and an experience people genuinely prefer.
Town’s value appears when it surfaces buried information, such as a missed dinner invitation, while leaving people to decide which suggested outcomes to accept.
The discussion widens from workplace assistance to agents that optimize benefits, time purchases, and manage information with less human mediation.
The central wager is that automated attention will weaken supply aggregation and force marketplaces, businesses, and trusted assistants to compete differently.
Weekly email
State of the Pod, every Monday
The week's most-discussed names and stories from podcasts, with the moments behind them. Free, five minutes, unsubscribe in one click.