When you purchase through links on our site, we may earn an affiliate commission.Heres how it works.
The company hopes its new tool will help prevent misinformation, as well as ensure proper attribution.
ExistingAImodels already generate text based on probabilities, or guessing which word comes next.
Image Credit: Geralt / Pixabay
The company also noted thoroughly rewriting a response can greatly reduce detector confidence scores.
Instead, it has been exploring its own watermarking technology.
