AI验证成本降低导致风险承担者减少
"When liability exposure approaches 0—when no one pays the price for unverified failures—verificatio...
朋友， Naval 提了一个很关键的观点：AI让容易验证的事变便宜了，但难验证的事还是难。现在没人为没验证的失败买单，导致验证预算缩水，大家都在用没监控的代理，这很危险。
当AI大幅降低易于验证任务的执行成本时，验证成为瓶颈。无人为未验证失败承担代价时，验证预算会崩溃。部署者将不受监控的代理涌入“失控风险区”，自动化收益被捕获，而灾难性风险被社会化。
"When liability exposure approaches 0—when no one pays the price for unverified failures—verificatio...
"When liability exposure approaches 0—when no one pays the price for unverified failures—verification budgets collapse...Deployers flood the Runaway Risk Zone with unmonitored agents, capturing the gains of automation while socializing catastrophic risks." Christian Catalini @ccatalini 1/ When we looked at the economics of AGI, the key policy challenge was immediately clear: AI drastically lowers the cost of execution for anything easy to verify. For everything else, verification is the bottleneck. 🔗 View Quoted Tweet 💬 53 🔄 18 ❤️ 279 👀 69287 📊 65 ⚡