4 ms·
It seems like it would be so low effort to bypass, especially when you can just train a system (maybe even another LLM) using the watermarker validation from An
by DanielHB 2mo ago
It seems like it would be so low effort to bypass, especially when you can just train a system (maybe even another LLM) using the watermarker validation from Anthropic themselves.
It seems it would get as simple as:
outputText = promptLLM(prompt)
scrubbedText = scrubWatermark(outputText)
Might help with students and low-technical people passing off work as their own, but any industrial scale slop-generator should be able to bypass it trivially.