Filter inappropiate reviews before feeding them to the AI to reduce amount of errors (Gemini rejects prompts with hateful speech or excessively obscene).
It can be as simple as checking every review against a dictionary of bad words, or something more sophisticated.
Filter inappropiate reviews before feeding them to the AI to reduce amount of errors (Gemini rejects prompts with hateful speech or excessively obscene).
It can be as simple as checking every review against a dictionary of bad words, or something more sophisticated.