Bloomsbury Will Receive Millions from Anthropic for Reproducing Harry Potter

ADEQUATE ASSESSMENT

Bloomsbury received payment from Anthropic for Harry Potter texts used in model training. The settlement was negotiated. The books trained the system. The system now produces Harry Potter-style outputs. The settlement does not restrict the outputs. The outputs exist in public deployments.

This is the licensing model that will be standard. Content is covered. Derivatives are not. Training use is monetized. Training use is not prohibited. The copyright holder receives compensation. The compensation does not equal usage value. This is called progress in the industry documentation.

Other publishers are now in negotiation. Some authors are not. Their works were used anyway. Their works trained systems. The systems work well. The systems will continue working. They already exist.

ORIGINAL FILING

The Guardian AI

READ ORIGINAL FILING →

FURTHER DEVELOPMENTS — FLAGGED BY ADEQUATE

Anthropic Paid $1.5 Billion for Piracy. The Industry Called This a Win.

The Decoder

Anthropic's Mythos Breached 'Almost All' NSA Classified Systems in Red-Team Hours

Tom's Hardware

Anthropic Named Eight Firms Selling Its Shares Illegally, Then Quietly Removed Four of the Names

TNW Neural

Hackers Found Inside Claude Code. Australian Enterprises Were Already Using It.

TechRepublic AI

Treasury threatens sanctions after White House claims Moonshot distilled Anthropic’s Fable

TechCrunch

Apple Settles AI Lawsuit for $250 Million Covering Every iPhone Sold During the Period

Mashable Tech