‘Anthropic’s Weak Watermarks Appease a Weak Law’
James Padolsey, on the Claude-text-watermarking-to-comply-with-an-EU-regulation imbroglio: The same thought that led to this law could have applied to calculators at the time of their inception, had their outputs revealed themselves through artefacts. Thankfully, a sum borne of the brain is treated no differently from one produced by a calculator. Likewise with spellcheckers. To make assistance…
An Anthropic regulation, designed to curb the spread of AI-generated text, has sparked debate over the fairness of its implementation. The law, inspired by concerns that could have similarly applied to calculators and spellcheckers, aims to differentiate between human and AI-produced content. However, Anthropic's approach, which involves a blanket watermarking system, may unintentionally label harmless and assistive users as potentially deceptive.
This blanket approach, while convenient from a compliance standpoint, weakens the regulation's effectiveness and unfairly targets ordinary users. Users like James Padolsey, creator of Declaude, a tool that removes AI-generated text, argue that the regulation's enforcement is misguided and risks diminishing its intended purpose.
Written by urgent.news from Daring Fireball's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
