I tried typing "propaganda" ten times with 100% accuracy. When it's preceded with "[left] wing", the accuracy drops to 0%.
29 karma · joined October 23, 2019
I tried typing "propaganda" ten times with 100% accuracy. When it's preceded with "[left] wing", the accuracy drops to 0%.
Picked at random: "right wing propaganda" isn't severe enough a phrase that I instinctively know I'm hitting a swear filter. After swiping the last word iOS goes back and changes the second word to "Wong". In order to understand what was happening and then fix it, I had to backspace (deleting the whole phrase), re-swipe both words, backspace, re-swipe both words again, double tap the incorrect word, backspace, type the word, tap to return to the end of the sentence, and hit space. All to fix one letter.
If this was really about protecting people from accidentally typing these that might land them in hot water, why wouldn't iOS at least offer the word it knew I wanted in the suggestion bar?
IMO: if your goal is to attract users or to make money, you have too many strong free competitors to be using ads like this, especially if your audience read Hacker News. It's not the right business model. Any amount of time spent making these ads 'smoother' is better spent on the product, because you won't earn enough ad revenue to make it worth it.
If you need revenue to pay for hosting, you'd be better off making it open source or self-hostable.
If your goal is getting feedback, you have to recognise that people are doing it for you, not for them. Don't make them pay you to do it.
Congrats on building this for yourself, I can certainly see the benefit of building writing tools to suit your own needs.
I imagine any potential user of a tool in this state would know what be wary of, but I do worry a little that LLMs have been trained on bad health advice or misinfo, and there will be very limited training data from real patient diagnoses. Even if dangerous data is minuscule in training terms, can good prompting de-risk dangerously bad advice?
However, for younger kids (say, 5-11) I think a separate environment is the best solution. Up to a certain point, parents have more influence on what their child does. For all its problems, most parents I know are comfortable with YouTube Kids because it's the lesser of two evils. I know that my child is going to participate in social media either way, so I want them to do it in a sandbox until the day I don't get a say.
I also don't think it has become much less common in communities where poverty levels have fallen. Anecdotally, my SO is part of a community originally displaced from India and now settled all over the world, where members have disproportionately become very wealthy. Bed sharing is absolutely the norm in those households, much to my distress when we visit her family!