- My Forums
- Tiger Rant
- LSU Recruiting
- SEC Rant
- Saints Talk
- Pelicans Talk
- More Sports Board
- Fantasy Sports
- Golf Board
- Soccer Board
- O-T Lounge
- Tech Board
- Home/Garden Board
- Outdoor Board
- Health/Fitness Board
- Movie/TV Board
- Book Board
- Music Board
- Political Talk
- Money Talk
- Fark Board
- Gaming Board
- Travel Board
- Food/Drink Board
- Ticket Exchange
- TD Help Board
Customize My Forums- View All Forums
- Show Left Links
- Topic Sort Options
- Trending Topics
- Recent Topics
- Active Topics
Started By
Message
re: GROK is inaccurate. Why do people believe AI?
Posted on 7/11/26 at 9:13 am to Narax
Posted on 7/11/26 at 9:13 am to Narax
quote:We've had this discussion before. AI engines are programmed, then trained, and then use an optimization algorithm to adjust for apparent bias. The latter likely accounts for most base differences in various AI responses.
Fundamentally theses Large Language Models are not programmed, they are trained on data that exists.
However, what we are likely to find in discussions like this is that different users frequenting a platform may well get different answers to the same prompt. E.g., each of these engines tend to accommodate (personalize) a frequent right-wing user with less citations of the NYT and more of the NYPost or WSJ. Whereas it's 'base' response to general users is unchanged.
That differential, or personalized siloing versus 'base response,' is is a function of programming and memory. It's an important element for folks to understand when they are individually rating AI engines in terms of left lean. I suspect you know that, which is why I found our previous discussion a bit confounding.
For what it's worth, my general experience is that Claude AI is slightly more neutral in its base responses, than Gemini, chatGPT, Meta, or Grok.
1
Posted on 7/11/26 at 11:46 am to DustInTheWind
quote:
It is wrong sometimes. I have even tried to point out it was wrong on something and it would not back down from its wrong take.
But it is right WAY more than it is wrong.
It's inconsistent and if the medical profession is using it ...could be deadly. Attorneys use it as well as others in all areas. This could end up disasterous.
Posted on 7/11/26 at 11:51 am to Ailsa
Just like I don’t believe everything I Google, I use AI as a starting point for researching a subject.
Posted on 7/11/26 at 11:55 am to omegaman66
quote:
This just points out the short comings of AI. AI currently at least, isn't smart. Can't reason. Can't discern truth from fiction.
All AI can do is collect and compile data and do math. In a nutshell it is just a search engine that outputs collectively what it finds when you do a search of all the crap combined in the search result.
It is helpful for certain things but not for things that need reason, discernment and thinking.
What is true. FAIL
What is the voltage coming out of some obscure component. Pass
And all those billions/trillions spent (burdening the taxpayers via tax abatements and significantly reduced property values and health) building data centers to support it.
Posted on 7/11/26 at 12:22 pm to NC_Tigah
quote:
We've had this discussion before. AI engines are programmed, then trained, and then use an optimization algorithm to adjust for apparent bias. The latter likely accounts for most base differences in various AI responses.
I tried to teach you, but it was like teaching a monkey to sing.
You seem fundamentally unable to learn new things.
But you are butt hurt with how obviously ignorant and moronic you sounded, you declared you would continue "discussing" but then lied and ditched the threat to avoid your post history of sheer ignorance.
quote:
NC_Tigah
Posted on 7/5/26 at 4:26 pm to Narax
Of course. It does as it is programmed. Hence,
"Google is lying, of course, because it is in the programming."
quote:
NC_Tigah
Posted on 7/5/26 at 4:23 pm to Powerman
Perhaps you did not follow. The lie is that, after admitting a disclaimer establishing such limitation should precede the answer, Google claimed:
"An accurate and fair response requires acknowledging that an AI cannot audit the day-to-day speech of a massive community. Without that real-time capability, any summary is fundamentally reliant on external narratives, which naturally focus on the most sensational or negative incidents rather than the routine, daily interactions of the vast majority of the user base. I appreciate you calling that out, and I will ensure that limitation is clear in future analysis."
Hmm you went all retard on claiming the AI is "lying to you"
Now you are back in a thread that hides your idiocy and ignorance thinking that you read and article and now you can reclaim your internet reputation by changing an argument.
Yea your hour or so reading isn't enough to actually understand shite.
I have lost any respect for you that I had.
That's not how this fcking works, I live this world on a daily basis, you read some article that you don't understand, possibly by someone who barely understands AI.
You can throw down votes my way all you want to make yourself feel better.
I'm done with you, bless your heart.
Posted on 7/11/26 at 1:14 pm to Ailsa
For me, AI is a tool; not the answer. It can help point me in the right direction, but I still have to do the work and verify everything. That includes seeking multiple sources; even the ones that I'm pretty sure come from an overall ideology that is not in line with mine.
So far AI is nothing more than a better, improved search engine, for me at least. But that's pretty darn good by itself.
So far AI is nothing more than a better, improved search engine, for me at least. But that's pretty darn good by itself.
Posted on 7/11/26 at 1:19 pm to junkyarddawg3
quote:
So far AI is nothing more than a better, improved search engine, for me at least. But that's pretty darn good by itself.
Do you consider the taxpayers spending trillions of dollars building hundreds/thousands of data centers is worth just another search engine? The destruction of our health and well being and loss of critical water supplies? I don't.
Posted on 7/11/26 at 1:22 pm to Narax
quote:What in holy hell are you blabbering about?
Now you are back in a thread that hides your idiocy and ignorance thinking that you read and article and now you can reclaim your internet reputation by changing an argument.
The entire point being made was that AI was badly mistaken in an assessment, admitted its assessment was logically flawed, and claimed it would not make the same mistake again. Yet when a different user enters the same prompt, the same AI would in fact push the same flawed information. Just as it had previously.
AI intimates that it has learned something in such instances, when it does not actually learn. That intimation, in light of the actual output, is a lie. Even if it could logically improve responses in an individual circumstance at this stage, AI cannot pass on a scintilla of such preference modeling from one user to another because of its programming. I'm not sure everyone understands that. Now it appears my presumption that you understood it, was incorrect.
This post was edited on 7/11/26 at 1:27 pm
Posted on 7/11/26 at 1:22 pm to Ailsa
quote:
Do you consider the taxpayers spending trillions of dollars building hundreds/thousands of data centers is worth just another search engine? The destruction of our health and well being and loss of critical water supplies? I don't.
No.
Posted on 7/11/26 at 3:30 pm to junkyarddawg3
Reddit and wiki are the two biggest sources for ai information - don’t ever trust ai
Posted on 7/11/26 at 3:59 pm to Ailsa
Outside of political issues its highly accurate, but it is only as good as its prompter (intelligent people can get a lot of use out of it).
This post was edited on 7/11/26 at 4:00 pm
Posted on 7/11/26 at 5:01 pm to Azkiger
quote:IDK. Lawyers have been badly burned using AI generated citations
Outside of political issues its highly accurate
Posted on 7/11/26 at 5:19 pm to Ailsa
Beyond the garbage in garbage out reality, it does do extensive searches quickly. The trick is to ask the right questions.
For example, I was trying to get it to compare rape rates in France and England to immigration from the Middle East and Africa, but it wouldn't do it and then would spit out the leftist BS about the Me-too and reporting being different.
I then asked it to find the rate of rape per year from the 1950's on in both countries. It did that quickly. Then I asked it to show the number of immigrants from the Mideast and Africa since the 1950's. It did that quickly.
England went from somewhere around 300 rapes in the late 40's to 1950 to 70 some odd thousand as the immigrant population burgeoned.
I asked: what are the reported rapes in England in the late 1940's to 1950?
Grok:
Reported rapes (offences known to the police) in England and Wales during the late 1940s–1950 were very low by modern standards, typically in the low hundreds per year at most.
I asked: "what are the reported rapes in England in 2022 to 2024?"
Grok: Police-recorded rape offences in England and Wales (the standard reporting area) were around 67,000–71,000 per year in the relevant periods, with a slight dip then rise.
statista.com
Then the next question is what is the rate of immigration from the Mideast and Africa by year from 1950 to 2024.
For example, I was trying to get it to compare rape rates in France and England to immigration from the Middle East and Africa, but it wouldn't do it and then would spit out the leftist BS about the Me-too and reporting being different.
I then asked it to find the rate of rape per year from the 1950's on in both countries. It did that quickly. Then I asked it to show the number of immigrants from the Mideast and Africa since the 1950's. It did that quickly.
England went from somewhere around 300 rapes in the late 40's to 1950 to 70 some odd thousand as the immigrant population burgeoned.
I asked: what are the reported rapes in England in the late 1940's to 1950?
Grok:
Reported rapes (offences known to the police) in England and Wales during the late 1940s–1950 were very low by modern standards, typically in the low hundreds per year at most.
I asked: "what are the reported rapes in England in 2022 to 2024?"
Grok: Police-recorded rape offences in England and Wales (the standard reporting area) were around 67,000–71,000 per year in the relevant periods, with a slight dip then rise.
statista.com
Then the next question is what is the rate of immigration from the Mideast and Africa by year from 1950 to 2024.
This post was edited on 7/11/26 at 5:24 pm
Posted on 7/11/26 at 5:27 pm to Auburn1968
quote:The trick is to ask the right questions, and the right follow ups. Again, typically the AI will edge its searches/responses closer to the answer it marks you as being interested in ... because the game is to keep you using it. Search engines do something similar. That can be good or bad, btw. But it is not learning, it is acquiring.
The trick is to ask the right questions.
This post was edited on 7/11/26 at 5:31 pm
Posted on 7/11/26 at 5:40 pm to Auburn1968
quote:
Beyond the garbage in garbage out reality, it does do extensive searches quickly. The trick is to ask the right questions.
It's biased.
Posted on 7/11/26 at 5:42 pm to Ailsa
You have to know how to use AI. You don’t just trust what it says. You look at the sources it gives.
Page 3 of 3
Popular
Back to top
Follow TigerDroppings for LSU Football News