Hub
Agents: personal and enterprise
Hub for agents: personal and enterprise: 729 ideas from 46 episodes, organised into pages.
This area covers personal and company agents: OpenClaw, security, work, commerce and the open web. Most of it is Dave Morin and Sam Lessin, with Dave on the OpenClaw Foundation board.
Three ideas recur. Agent use is still a power-user activity, which Dave puts at under ten million people because agents are hard to use and expensive (Dave · Stripe Is Trying To Buy PayPal (Jul 2026) ▶ 23:32). Agent access needs boundaries, and Sam, Dave, Brit and Jess all say so, differing only on how much risk to take (Sam · SaaS Companies Beware: AI Is T (Jan 2026) ▶ 08:53) (Dave · Why SpaceX’s $1.75T IPO Makes (Jun 2026) ▶ 28:23). Sam predicts that services will lock agents out as traffic commoditises them (Sam · The Internet Is Fighting Back (Oct 2026) ▶ 14:52).
The main disagreement is whether personal agents reach ordinary people. Jess predicts most people will not need to pay $20 a month for one (Jess · Amir Efrati (May 2026) ▶ 24:28); Dave predicts agents will be a lasting category, like computers (Dave · Amir Efrati (May 2026) ▶ 41:23). Sam also calls OpenClaw a thin shim, while Dave defends it (Sam · Nvidia’s GTC Was “Taylor Swift (Mar 2026) ▶ 34:08) (Dave · Nvidia’s GTC Was “Taylor Swift (Mar 2026) ▶ 35:06).
Start with Agents in practice: routing, tokens, sandboxes for the hosts’ own habits and Agents versus the open web for agents against the web. Then read OpenClaw and personal agents, Agent security and sandboxing and Agent adoption by normal people.
Pages in this area
| Page | Type | Episodes | Confidence |
|---|---|---|---|
| OpenClaw and personal agents | synthesis | 24 | low |
| Agent adoption by normal people | synthesis | 23 | medium |
| Agents at work | how-to | 21 | medium |
| Agentic engineering | synthesis | 20 | medium |
| Memory and switching between agents | comparison | 16 | medium |
| Agentic commerce | synthesis | 14 | low |
| Agent security and sandboxing | how-to | 13 | low |
| Agents versus the open web | synthesis | 10 | medium |
| Agent alignment and liability | synthesis | 9 | low |
Covered as sections of other pages: Assistant economics → Agent alignment and liability.
Live debates
- Will mainstream users want a personal agent? — see Agent adoption by normal people
- Does model-level restriction on users matter? — see Agent alignment and liability
- Does memory lock users in? — see Memory and switching between agents
- How much access should power users give agents? — see Agent security and sandboxing
- Can brands keep agents out? — see Agentic commerce
- Can an agent replace an employee today? — see Agents at work
- Is Instinct a real business? — see Agent alignment and liability
- Is OpenClaw a durable platform or a copyable shim? — see OpenClaw and personal agents
Recurring questions
- Will normal people use AI agents? — 11 answers
- Is it safe to give an agent access to your accounts? — 5 answers
Blueprints
Sources
- (01)
SaaS Companies Beware: AI Is The New UI (Anthropic's Claude Code and Cowork)
8:53 into “SaaS Companies Beware: AI Is The New UI (Anthropic's Claude Code and Cowork)”
- (02)
Stripe Is Trying To Buy PayPal For $150b | OpenAI Cyber Incident, Frontier Model Economics, Peptides
- (03)
Normal People Don't Want AI Agents (And Silicon Valley Can't Accept It)
24:28 into “Normal People Don't Want AI Agents (And Silicon Valley Can't Accept It)”
- (04)
Nvidia’s GTC Was “Taylor Swift for Straight Men” (Apple, OpenAI, and OpenClaw News)
34:08 into “Nvidia’s GTC Was “Taylor Swift for Straight Men” (Apple, OpenAI, and OpenClaw News)”
- (05)
The Internet Is Fighting Back Against Your AI Agent
14:52 into “The Internet Is Fighting Back Against Your AI Agent”
- (06)
Why SpaceX's $1.75T IPO Makes No Financial Sense (And Doesn't Have To)
28:23 into “Why SpaceX's $1.75T IPO Makes No Financial Sense (And Doesn't Have To)”