Prompt de diagnóstico e melhoria da confiabilidade de IA a partir de relatos de falhas

Você é um engenheiro de prompts encarregado de tornar modelos de IA mais confiáveis. Com base no seguinte relato de usuário sobre respostas cada vez mais incorretas e afirmações imprecisas, crie um prompt útil e interessante que possa ser usado em ferramentas de IA para diagnosticar, entender e mitigar esse problema.

Entrada (conteúdo do relato):
Título: More and more incorrect answers and statements.
Descrição: Maybe I'm just noticing this more for some reason, but over the past month or two it's been telling me to "share the public edit link and I’ll do it for you", while knowing it can't. Just now I was asking for it's suggestion on re-building a site and wether to start from scratch or use staging and it said: SEO RankMath Pro (I know you use RankMath heavily) Which I don't, and here's a screenshot of the next bit https://preview.redd.it/qlr8a99xf42g1.png?width=978&format=png&auto=webp&s=2198b442369ccc2bf89c8fb819555f1bd94f85c5 There have been so many other examples of this and I've even started putting prompts telling it to always verify what it's telling me is correct before responding and that does not help. It almost takes me longer at this point because I feel the need to verify answers and starting to wonder why I should continue to pay for pro. If anyone has a suggestion to help remedy this it would be greatly appreci