Posts: 185
Threads: 17
Joined: Feb 2021
(03-06-2026, 05:10 PM)prostatenipple Wrote: Although I'm very distrustful of "AI" (because it's just predictive text with a very large dataset, and I work in IT and I see what lies it gives)
Fully agree... Don't make any important decisions solely based on an AI conversation.
I work with AI ( mostly llama based models ) and there are instances where the answer AI gives you completely contradicts previous prompt. It is a great resource for researching a topic, or finding evidence to support your hypothesis,
but far from perfect.
Posts: 403
Threads: 6
Joined: Sep 2020
If an AI doesn't cite a source, its worthless. even then, it just helped you find the source. It can cite sources that tell you the opposite of what the AI is saying. its the fruit of a poisoned tree. You would be best off disregarding it entirely because it can seep between your defenses, any little bit of intellectual laziness and it can feed you correct sounding falsehoods.
Posts: 3,262
Threads: 16
Joined: Mar 2019
I agree with Dru on using AI especially on NBE matters... Firstly, its not an intelligence, its a text guessing machine, secondly without sources its useless information. How can you know if its feeding you mis/disinformation or just what you want to hear? AI chatbots are notorious for giving exactly what the prompts asks, where's the fact check on what its telling?
Posts: 185
Threads: 17
Joined: Feb 2021
(05-06-2026, 09:19 AM)DruLactin Wrote: If an AI doesn't cite a source, its worthless.
When copied the reference links get omitted. Many of the references to research studies, along with references to this site and reddit. I fully agree the information it provides is only as accurate as it's source.