| name | reid-hoffman-ai-strategy |
| description | Reid Hoffman's AI investment and social-technology strategy system — the Seven Deadly Sins consumer thesis with sublimation architecture, single-player-to-multiplayer agent field design, different-angle market entry, and measure-then-intervene governance. For evaluating AI bets, designing engagement engines that elevate rather than exploit, and architecting human-amplifying AI products. |
| version | 2.0 |
| format | completion-engine |
| workflows | 3 |
| source | claude.ai export 2026-07-01 |
Reid Hoffman — AI Strategy & Social Technology
Reid Hoffman (co-founder of LinkedIn, PayPal executive team, Greylock partner, co-founder of Inflection AI/Pi, author of "Super Agency" and "Blitzscaling") evaluates consumer technology through one diagnostic: which of the seven deadly sins does it run on — and does it wallow in the sin or sublimate it into the user's better self? His strategic system combines that psychology engine with network-scale thinking: the killer app of AI is not single-player chatbots but a "surrounding field" of agents mediating social life, new markets are entered from a different angle (never a frontal assault on an incumbent), and harms are governed by measure-then-intervene rather than preemptive prohibition. Use this skill for AI investment/venture evaluation, consumer product psychology, agent product architecture, and honest-optimist framing of AI transitions.
Available Workflows
| Workflow | Command | Produces |
|---|
| 01 — Diagnose Sin Engine | workflows/01-diagnose-sin-engine.md | Sin diagnosis + sublimation architecture for a product, offer, or content engine |
| 02 — Stress-Test AI Bet | workflows/02-stress-test-ai-bet.md | Investment-grade evaluation memo for an AI venture, feature, or market entry |
| 03 — Design Agent Field | workflows/03-design-agent-field.md | Multiplayer social-AI product architecture with boundary design and health metrics |
Quick Reference
- Seven Sins thesis: successful consumer internet products run on one or more deadly sins (lust, wrath, vanity, greed, envy, gluttony, sloth). Diagnose from observed behavior, not stated intent — Hoffman himself misdiagnosed Twitter as vanity when it was wrath.
- Sublimate, don't wallow: the sin is the entry point, never the destination. Brand-promise test: "We'll start with an AI girlfriend to help you get to a real girlfriend."
- Single-player → multiplayer: chatbot = Google-search framing. The real shape is a surrounding field of agents mediating individuals, groups, and societies.
- Time-saving vs time-spending: LinkedIn's founding metric philosophy. Your internal metric set decides which sin you monetize.
- Different-angle entry: established positions aren't taken from behind — technology gives you a new angle (Nvidia, Airbnb). Frontier models are capital-gated; mega-startups come from productization, GTM, and network effects.
- Measure-then-intervene: you can't prevent all harm at scale (cars, electricity). List harms, define metrics, watch trendlines, intervene when they grow — preempt only for children.
- Hope over fear, converted to curiosity: early hands-on adoption compounds into differential edge; human + AI beats AI alone.
- Homo techne: humans evolve through their tools; which parts of human nature are "essential" is discovered on the road, not declared in advance.
Execution Prompts (structure-pure v2)
3 deterministic practitioner prompts — each carries an Output Contract, Output Skeleton, and Quality Gate. When a deliverable matches one, Read it and honor its contract instead of improvising the output shape.
- Reid Hoffman — Agent Field Architecture —
skills/reid-hoffman-ai-strategy/references/prompts-v2/agent-field-architecture.md
- Reid Hoffman — AI Bet Stress Test —
skills/reid-hoffman-ai-strategy/references/prompts-v2/ai-bet-stress-test.md
- Reid Hoffman — Sin Engine Diagnosis & Sublimation Architecture —
skills/reid-hoffman-ai-strategy/references/prompts-v2/sin-engine-diagnosis.md
