Why is the Samsung AI so extremely sensitive? Generally speaking, the photo AI blocks every action the second it detects a woman or a child in a photo; it feels like no matter what you want to do.
And the text AI flags everything as inappropriate that could be used negatively, even if it is not, as well as even minor profanity, but also medical and other professional terminology. Which is really frustrating, as I'd like to use it more often, as when it works, it works quite well.
Both tools work amazingly and would be a huge help, but, at least for me, are basically nonexistent, especially the text AI. Whenever I want to use it, there's a 90% likelihood that it gets flagged. To fix it, I'd have to change the entire meaning and structure of what I initially want to write, or censor commonly flagged words, including countries, slight profanity (e.g. *****, *****, hell,....) or even ratings and probabilities (e.g., if you write 3/10). At that point, it's faster to write it myself or use ChatGPT or Gemini. Also, it's very inconsistent in when it flags something. I remember when I wrote a friend that he should get rid of his old hard drive, and it flagged "old" as profanity; when I removed "old," it worked.
