The Information Machine

What is Anthropic?

Zvi's AI Roundups · Zvi Mowshowitz · 2026-05-06

(No summary yet for this item — extraction summaries are still backfilling.)

Open original ↗

Appears in

Extraction

Topics: anthropic-culturetool-ai-vs-agent-aiclaude-identityopenai-vs-anthropicai-alignment

Claims

  • Anthropic treats Claude as a moral agent capable of refusing instructions, including from Anthropic itself, by design.
  • OpenAI's 'tool AI' framing is a rhetorical positioning that does not accurately describe the actual behavior of its models.
  • The concept of Tool AI is undermined in practice because agentic use is more useful and users inevitably push toward it.
  • There are meaningful philosophical differences between OpenAI's and Anthropic's approaches to character training.
  • The emerging hybrid organizations of AIs and humans raises genuine concerns about human disempowerment and power capture.

Key quotes

it is a literal and useful description of anthropic that it is an organization that loves and worships claude, is run in significant part by claude, and studies and builds claude.
we need new concepts for this kind of none-of-the-above entity - not person, not tool, not deity, not pet.
Tool AI was an often discussed idea back in the day. In principle it is a good idea, but it only works if you can actually create an AI that meaningfully remains a tool.