In a new podcast episode titled "Why 'Virtual Intelligence'?," Christopher Horrocks makes a compelling case for a more precise term to describe large language models and similar systems. The episode argues that common language around artificial intelligence conflates fluency with true agency, leading to confusion about where accountability lies.
Horrocks reads from his essay "Why 'Virtual Intelligence'? Naming, Agency, and Accountability in the Age of Large Language Models," the second in a series. He explains why existing terminology fails to capture the nature of these systems, what genuine agency requires, and why the distinction between merely fluent outputs and self-governance determines who—or what—is responsible for actions.
The episode is part of Horrocks' podcast series on Virtual Intelligence, which explores these themes in depth. Listeners are invited to read the full essay and join the discussion at chorrocks.substack.com.