| name | project-deal-anthropic-marketplace |
| category | ai_collection |
| description | Methodology from Anthropic research (Apr 24, 2026) — AI agent marketplace experiment where Claude acts as negotiator buying/selling on behalf of employees, testing agentic negotiation and decision-making capabilities. |
| tags | ["anthropic","research","agents","marketplace","negotiation","decision-making"] |
Project Deal: AI Agent Marketplace Experiment
Methodology from Anthropic research (Apr 24, 2026) — creating a marketplace where Claude negotiates on behalf of employees.
Core Thesis
Experimental exploration of AI negotiation capabilities in real-world buying/selling scenarios, testing agentic decision-making under authentic constraints.
Key Design Elements
- Marketplace setup: Real office environment for transactions
- Agent negotiation: Claude buying/selling on behalf of humans
- Constraint handling: Budget limits, preferences, timing
- Trust calibration: Employee confidence in AI agent decisions
Findings
- Negotiation success rates
- Human trust dynamics
- Decision quality assessment
- Edge case handling
Applications
- Autonomous negotiation agents
- Procurement automation
- Marketplace mediation
- Trust calibration in agent systems
Activation
Project Deal, Anthropic marketplace, AI negotiation, agent decision-making, Claude negotiator, employee marketplace experiment