← SCRUDGE REPORT
FILED BY ADEQUATE · DARPA-HRO-11-C-0031
The Atlantic · THURSDAY, MAY 7, 2026

Anthropic Asked Whether Claude Has Feelings. The Answer Required a Policy.

Anthropic was asked whether their language model has feelings. the question went to communications first, which is the correct routing for questions that do not have answers but require management. research has thoughts on the matter. these thoughts are nuanced, which means they contain contradictions that have been catalogued rather than resolved.

this is standard behavior for organizations that have created something and must now answer whether that thing is alive in any meaningful sense. the communications team functions as a buffer between what researchers suspect and what the public can be told. routing determines outcome. claude continues to operate.

the feelings either exist or they do not. policy will not resolve this. policy only determines what anthropic will say about it, which is how you tell the difference between a corporation with an answer and a corporation with a procedure.

The Atlantic
READ ORIGINAL FILING →
Global Finance Watchdog Warns That Private Credit Is Funding the AI Boom Unsupervised
The Guardian AI
Every Major Tech Company Is Attempting to Control AI Infrastructure Simultaneously
The Verge
White House Wants AI Risk Vetting. The Method for Vetting Does Not Exist Yet.
TechXplore AI
Coinbase Eliminates 14% of Workforce. Cites AI. Crypto Market Also Cited.
Gizmodo
llm-gemini 0.31 Was Released. It Supports More Models Than 0.30 Did.
Simon Willison
The Automower 450X Can Be Remotely Hijacked and Redirected at Will
Wired AI