What Is A Meat Proxy?

A “meat proxy” is a deliberately unflattering term for someone who passes on AI-generated work without properly reading, understanding or checking it first. The human is technically involved, but largely acts as the middleman between the AI and whoever receives the output.

Using AI doesn’t automatically make someone a meat proxy. You might ask it to improve something you’ve written, summarise information or challenge an argument. If you read what comes back, question it, make changes and take responsibility for the result, you’re still applying human judgement.

The distinction becomes interesting when we talk about keeping a “human in the loop”. Having a person somewhere in an AI-supported process doesn’t necessarily provide meaningful oversight if their only contribution is clicking approve, copy or send.

The point of having a human involved is presumably that they bring something the AI doesn’t: experience, context and the judgement to recognise when something doesn’t look right. If they’re simply passing on whatever the AI produces, that safeguard starts to become fairly meaningless, even if an organisation can technically say there was a person involved.

Perhaps that’s why “meat proxy” is such a wonderfully horrible term. It describes the difference between someone genuinely working with AI and someone simply becoming the final step between the AI and everybody else.

As AI becomes part of everyday work, perhaps the useful question isn’t simply whether a human was involved, but what they actually added.