Reasoning: Our AI model only takes sentence by sentence, thus allowing the users to see the surrounding sentences may clear misunderstandings. E.g. There might be a sentence "Continued use of the service signifies acceptance of the changed terms and services" which the model would mark as dangerous. However, there might be an extra following sentence of "We will notify you of the change in terms and services over a reasonable amount of time", making the term less dangerous.
thegrimbee
commented
1 month ago
Reasoning: Our AI model only takes sentence by sentence, thus allowing the users to see the surrounding sentences may clear misunderstandings. E.g. There might be a sentence "Continued use of the service signifies acceptance of the changed terms and services" which the model would mark as dangerous. However, there might be an extra following sentence of "We will notify you of the change in terms and services over a reasonable amount of time", making the term less dangerous.