Pinned
Today, we're kicking off the first phase of the research preview for Model Hardware Standard (MHS): a new standard for AI agents to safely operate physical equipment in scientific research and advanced manufacturing. Read more: anthropic.com/news/model-har…- we have a decent proxy: self interest + free market + rule of law. however, digital intelligence breaks the balance so hard that new balance is even hard to imagine ...A modest proposal: we just need to design an AI benchmark specified in such a way that the only way to cheat is to measurably improve the world. In homage to reverse psychology let's call this strategy "reverse alignment"
