Welcome to DU!
The truly grassroots left-of-center political community where regular people, not algorithms, drive the discussions and set the standards.
Join the community:
Create a free account
Support DU (and get rid of ads!):
Become a Star Member
Latest Breaking News
Editorials & Other Articles
General Discussion
The DU Lounge
All Forums
Issue Forums
Culture Forums
Alliance Forums
Region Forums
Support Forums
Help & Search
General Discussion
In reply to the discussion: How do you use ChatGPT and others like it in your daily life? [View all]highplainsdem
(59,280 posts)36. Of all the reasons given for LLM hallucinations, guardrails have never been mentioned in all the
articles I read about them. Guardrails are generally believed to reduce hallucinatiins.
And open source LLMs have also been trained on stolen intellectual property, so using them is unethical.
Edit history
Please sign in to view edit histories.
Recommendations
2 members have recommended this reply (displayed in chronological order):
59 replies
= new reply since forum marked as read
Highlight:
NoneDon't highlight anything
5 newestHighlight 5 most recent replies
RecommendedHighlight replies with 5 or more recommendations
I've found that for the little things I've listed it's usually accurate. I treat it...
LAS14
Friday
#14
Please don't assume any answers from it or the others are correct. Even on simple things where even the online news
hlthe2b
Friday
#10
So you're fine with it having been trained on stolen intellectual property? You don't care about
highplainsdem
Friday
#28
It's pretty easy to proof the questions I put to it in categories 1 and 2. As for...
LAS14
Friday
#53
Never used Chat GPT. Have used AI to figure out how to prepare a piece of
allegorical oracle
Friday
#22
What should you call it? An illegally trained plagiarism tool that dumbs you down while harming the environment.
highplainsdem
Friday
#25
LLMs like GPT are restricted by "Safety guardrails" so they don't offend anyone. That's the cause of most hallucinations
Hellbound Hellhound
Friday
#31
Of all the reasons given for LLM hallucinations, guardrails have never been mentioned in all the
highplainsdem
Friday
#36
No need to get personal here, mate. My principles are just fine, exactly where they belong.
Hellbound Hellhound
Friday
#43
The AI companies knew they were guilty of IP theft and copyright infringement. There's plenty of
highplainsdem
Friday
#48
Other than a handful of times out of shear curiosity, I haven't touched it.
Tommy Carcetti
Friday
#50