3 ms·
I'm kind of annoyed by ChatGPT claiming - in an authorative tone - things that are wrong. This particular model should be trained to be humble and admit when it
by Tepix 4y ago
I'm kind of annoyed by ChatGPT claiming - in an authorative tone - things that are wrong. This particular model should be trained to be humble and admit when it doesn't know something instead of making stuff up.
- seydor 4y agomaybe this is also Dangerous?
- raldi 4y agoIndeed: https://mobile.twitter.com/MikeSchiraldi/status/1587902753984024576 https://mobile.twitter.com/MikeSchiraldi/status/158790275398... https://mobile.twitter.com/MikeSchiraldi/status/1584431998252171264 https://mobile.twitter.com/MikeSchiraldi/status/158443199825... Edit: The new ChatGPT system does an excellent job with these prompts!