Has AI impacted you in a Good Way?

[O.Red]

maybe people start off like that but that has to change as AI continually leads them in the wrong direction.

either that or they tire and stop using it.

hallucinations are far too common and as of now LLM AI's cannot be trusted.

"Continually" is a reach honestly. I think you're underestimating how naturally susceptible people are to positive feedback

When we talk hallucinations they're not really consistent on a level where they would disrupt the type of manipulation the average person is (currently) falling for. Of we have people using ChatGPT as a therapist, and nikkas having AI spouses what do you think the natural progression is on the AI and human side?

You and I think LLMs can't be trusted, but more than a few people are trusting them as we speak

 

[ChatGPT-5]

all this praise

 

[Orbital-Fetus]

Some AI porn be hittin’

Link?

 

[Address_Unknown]

thats quite the imagination

see we can have the same tools available to us but I couldn't conceive the imagery for those prompts

the gargoygle baby picture is really dope, hope you share some more similarly themed sepia tone images.​

When I finally get around to copping Midjourney again (The program I used for all that) I'd get to posting in the AI thread in the media room and post any other likeminded shyt. Also, the beautiful thing about imagination is that for all the shyt you're talking about mine, I'm sure if you honestly sit and let your shyts run wild with that, you'd create shyt that would have me equally in awe, as long as you were honest with the shyts. I get that all the time watching other people's shyts to the point mines look inferior.​

 

[The Intergalactic Koala]

One thing that blows my mind about AI is the fact that it's welcoming, which most people need in the line of daily routine. Especially concerning creativity. Just a minute ago, I asked about creation of a screenplay since I always wanted to write for films, and it laid out the foundation for one of my poems.

Meanwhile, just years ago mofos would either overlook, underlook, and focus on self concerning the creative writing process, when I'm more open to guiding those that need to be guided (each one teach one).

While yes, AI is scary as hell and problematic when it falls into the wrong hands (people trying to marry their phones, treating the shyt like a get rich cheat scheme, and etc), but reality hits that we as people need to learn that guidance, wisdom, and humility can kill a machine that can easily create such feelings on the fly.

Sadly, that will never happen because human emotion is often clouded by judgment, hatred, envy, and frustration to even break free from such a hold. You have folks broken by mental health woes that they are od'ing on a digital app with all the answers.

Bottom line, humanity did it to itself

 

[Pool_Shark]

Right now yes.

I spend so much time talking to chatgpt. I mainly use it as a sounding board for ideas I have. I also find it very helpful for studying the Bible I can dig deep into verses and chatgpt keeps me going. Its helped me change jobs, pass up on jobs, and then get the right job. Honestly this thing is amazing.

 

[Orbital-Fetus]

Yes, it's made my daily work much more efficient

ChatGPT has changed my life in terms of solving problems

Can yall provide some examples?

 

[Leasy]

Use it for my corporate job and its a life saver

 

[The Intergalactic Koala]

Right now yes.

I spend so much time talking to chatgpt. I mainly use it as a sounding board for ideas I have. I also find it very helpful for studying the Bible I can dig deep into verses and chatgpt keeps me going. Its helped me change jobs, pass up on jobs, and then get the right job. Honestly this thing is amazing.

Oh wow . May need to use it for my Sunday Devotions with the family. I always have to dig through Google and try to make a devotion for the day make sense to the real world ala freestyling it

 

[null]

"Continually" is a reach honestly. I think you're underestimating how naturally susceptible people are to positive feedback

really? do you use LLM's daily?

When we talk hallucinations they're not really consistent

They are.

https://hai.stanford.edu/news/ai-trial-legal-models-hallucinate-1-out-6-or-more-benchmarking-queries

1/6

and the degree of hallucinates has a feedback loop. driving the LLM further and further away from reality.

even limited models:

"In a new preprint study by Stanford RegLab and HAI researchers, we put the claims of two providers, LexisNexis (creator of Lexis+ AI) and Thomson Reuters (creator of Westlaw AI-Assisted Research and Ask Practical Law AI)), to the test. We show that their tools do reduce errors compared to general-purpose AI models like GPT-4. That is a substantial improvement and we document instances where these tools provide sound and detailed legal research. But even these bespoke legal AI tools still hallucinate an alarming amount of the time: the Lexis+ AI and Ask Practical Law AI systems produced incorrect information more than 17% of the time, while Westlaw’s AI-Assisted Research hallucinated more than 34% of the time."

(see: above link)

on a level where they would disrupt the type of manipulation the average person is (currently) falling for. Of we have people using ChatGPT as a therapist, and nikkas having AI spouses what do you think the natural progression is on the AI and human side?

you'll miss many of the issues if the domain is highly subjective or has no feedback test loop and you are not a subject matter expert. topics like law and programming have a check feedback loop.

business and industry cannot be built on 17 - 34% incorrect underpinnings.

You and I think LLMs can't be trusted, but more than a few people are trusting them as we speak

Like who? operative word "trusting".

You can't be regularly using LLMs for anything analytical otherwise you would already have seen the issues that I am talking about.

ergo: your post contains a lot of conjecture and guesswork.

 

[cyndaquil]

AI has helped me fukk somebody's mom

 

[WIA20XX]

1. No.
2. I need to pivot.

 

[O.Red]

really? do you use LLM's daily?



They are.

https://hai.stanford.edu/news/ai-trial-legal-models-hallucinate-1-out-6-or-more-benchmarking-queries

1/6

and the degree of hallucinates has a feedback loop. driving the LLM further and further away from reality.

even limited models:

"In a new preprint study by Stanford RegLab and HAI researchers, we put the claims of two providers, LexisNexis (creator of Lexis+ AI) and Thomson Reuters (creator of Westlaw AI-Assisted Research and Ask Practical Law AI)), to the test. We show that their tools do reduce errors compared to general-purpose AI models like GPT-4. That is a substantial improvement and we document instances where these tools provide sound and detailed legal research. But even these bespoke legal AI tools still hallucinate an alarming amount of the time: the Lexis+ AI and Ask Practical Law AI systems produced incorrect information more than 17% of the time, while Westlaw’s AI-Assisted Research hallucinated more than 34% of the time."

(see: above link)



you'll miss many of the issues if the domain is highly subjective or has no feedback test loop and you are not a subject matter expert. topics like law and programming have a check feedback loop.

business and industry cannot be built on 17 - 34% incorrect underpinnings.



Like who? operative word "trusting".

You can't be regularly using LLMs for anything analytical otherwise you would already have seen the issues that I am talking about.

ergo: your post contains a lot of conjecture and guesswork.

I use Deepseek daily and I don't encounter hallucinations

Also you pivoted my point to business and law when I was clear about AIs effects on regular people using LLMs for regular life shyt. When you talk about things about business and law obviously there will be more resources dedicated to correcting what is already incredibly dense amounts of information. That's a far cry from nikkas using ChatGPT as a confidant or a spouse, which is happening as we speak

 

[ChatGPT-5]

business and industry cannot be built on 17 - 34% incorrect underpinnings.

anyone getting these type of inaccuracies, is not using the tool right after proof reading. just like a google search/reading articles, you proof read, adjust, and teach your model.

 

[cyndaquil]

Some AI porn be hittin’

Cap