zyruh

6 ms·

1.
Why is AI search still bad at trust and context?
2 points by zyruh 1y ago|1 comments
2.
by zyruh 1y ago
I’m curious what this community thinks: if an AI-powered search engine launched that significantly reduced hallucinations, improved transparency, and always cited its sources—would you at least give it a try and provide feedback? I’m buildi
3.
by zyruh 1y ago
What kind of AI product?
4.
If you could create any software product – what would it be?
5 points by zyruh 1y ago|4 comments
5.
by zyruh 1y ago
Right - that's what I'm thinking as well.
6.
Perplexity Comet – interested in your feedback
2 points by zyruh 1y ago|4 comments
7.
by zyruh 1y ago
Yes, and this is exactly why I'm creating a tool to address the slop and inaccuracy issues.
8.
by zyruh 1y ago
This is great - thank you!
9.
by zyruh 1y ago
Thank you for your comment.
10.
If You Could Fix One Thing About AI Search, What Would It Be?
1 points by zyruh 1y ago|7 comments
11.
HN vs. Reddit – Pros and Cons of Each
1 points by zyruh 1y ago|1 comments
12.
Improving AI Search to Reduce Hallucinations, Bias, and Memory Loss – Feedback?
1 points by zyruh 1y ago|1 comments
13.
Ask HN: What's Your Take on Perplexity AI?
2 points by zyruh 1y ago|2 comments
14.
Thoughts on ChatGPT-5? Especially Around Accuracy, Critical Thinking, and Memory
1 points by zyruh 1y ago|1 comments
15.
What's Your Favorite LLM –and Why?
5 points by zyruh 1y ago|3 comments
16.
by zyruh 1y ago
Makes sense. Is there another LLM you're using for this that works better than GPT?
17.
by zyruh 1y ago
Yeah, its irritating b/c I find myself wanting to trust - but its difficult due to its inconsistency.
18.
by zyruh 1y ago
No - just a frustrated consumer.
19.
by zyruh 1y ago
So, are you preferring a UX that feels more human?
20.
by zyruh 1y ago
The question is, "does it know there's a problem?"
21.
by zyruh 1y ago
So, what other LLMs does anyone recommend?
22.
by zyruh 1y ago
That's what happens when you're first to market.
23.
by zyruh 1y ago
LLMs need regular transparency - fact checking so the user can verify an validate accuracy.
24.
by zyruh 1y ago
Yes, I'm very much human...LoL.
25.
by zyruh 1y ago
I think we'll get there in time.
26.
by zyruh 1y ago
Yes, the LLMs need to be objective but in situations where its a subjective push back, the LLM would then need to take on a personality of its own.
27.
by zyruh 1y ago
Understood - thank you!
28.
by zyruh 1y ago
Interesting.
29.
by zyruh 1y ago
Great feedback!
30.
by zyruh 1y ago
In light of these issues, has anyone made a permanent switch to another LLM? If so, which one?

More ›