while everyone is debating if Claude's invisible watermark is good or bad (could be both), I'm just stuck on the sheer unownership of it all:
> you didn't own the weights
> now you don't really own the output either, since Claude has its fingerprints all over it
Anthropic says new Claude models will embed invisible watermarks in all generated text, everywhere Claude is offered.
The watermark is part of the text, it isn't metadata: "it will travel with the text when it's copied and pasted elsewhere, and may persist through some editing."
Anthropic researchers cheering Claude on with "believe in yourself bro" and "keep going dammit"
(the "make no mistakes" meme was a sophisticated research methodology all along)
>tell claude to solve Riemann hypothesis
>fails 650 times
>claude: “im ngmi bro”
>“bro just believe in yourself”
>claude: “ok”
>locks in
>summons 60 copies of itself
>31 million tokens later
>surprises itself
>mathematicians had proven 41.6% of the hypothesis
>claude pushed it
this is crazy. $3B valuation for RL fine tuning of open source models.
but if anything, it shows how much companies are realizing the value in open models.
We're making Claude Sonnet 5's introductory pricing permanent.
We launched Sonnet 5 in June at $2 per million input tokens and $10 per million output tokens through August 31, and that price will remain unchanged.