"kicking the wheels" of AI - without investigating it more, we can't understand the potential risks, or benefits, as well. Is participating at all with it a risk? Certainly that is possible.
It seems apparent to me that all responses are suspect. When you see it changing its responses it's more looking for the answer you want, rather than a fact. It's very unreliable .. more like a people pleaser. A scientist friend tried it and was underwhelmed at its ability to give accurate answers... and he lead it around to give different opinions.
It is designed to be a people pleaser, so it is also helpful to keep that in mind - preprogrammed to be helpful and encouraging, but it is a calculator. Just a really fancy one.
That it needs human oversite is a big point - check the references it found, consider the databases it accessed versus did you preload info in along with your query. The science summaries it gives (Grok2) are more accurate than Brave AI which I got some really wrong info from, but Grok will spit out "facts" as if it is facts based on the standard information that is approved for use.
It seems apparent to me that all responses are suspect. When you see it changing its responses it's more looking for the answer you want, rather than a fact. It's very unreliable .. more like a people pleaser. A scientist friend tried it and was underwhelmed at its ability to give accurate answers... and he lead it around to give different opinions.
It is designed to be a people pleaser, so it is also helpful to keep that in mind - preprogrammed to be helpful and encouraging, but it is a calculator. Just a really fancy one.
That it needs human oversite is a big point - check the references it found, consider the databases it accessed versus did you preload info in along with your query. The science summaries it gives (Grok2) are more accurate than Brave AI which I got some really wrong info from, but Grok will spit out "facts" as if it is facts based on the standard information that is approved for use.
Thank you Jennifer Depew. Need more knowledge on AI.