ChatGPT Do you also BLINDLY trust ChatGPT or AI?

[Rem]

https://www.msn.com/en-au/news/other/chatgpt-lied-to-us-furious-couple-blames-ai-for-missed-flight/ar-AA1Ljzjg

Unless you are god fucking forbid retarded
always do due diligence and fact check shit

Like serious to god
who the fuck decides to travel & based VISA/ESTA/eTA etc information on ChatGPT alone?

Take a deep breath
and take a look at this face
this is one of the leading face of retards out there

This applies to everything
including your code, your articles and everything inbetween
Yes, even your "AI images" could be straight up stolen

 

[Dopious]

Na, I always double check answers and text, heck I don`t even trust myself, hmm.

 

[Octavia]

Na, I always double check answers and text, heck I don`t even trust myself, hmm.

Shirt, bra, underwear, trousers, socks, shoes?

 

[Dopious]

Shirt, bra, underwear, trousers, socks, shoes?

Na, I work from home, big balls needs air FFS.

 

[Rem]

Na, I work from home, big balls needs air FFS.

birthday suit for work ?

 

[Dopious]

birthday suit for work ?

Yeah, no teams meetings for me, last one I got banned.

 

[polecat]

No I use it for quick facts or data checks to look for resources for further research

 

[tiiberius]

Yes, it's certainly getting to the point of blindly following anything it says.
More and more people falling for it.
I think that is the end game actually.

 

[p3n7357]

Hell no even my code needs to be triple checked it usually gives out very bad ones

 

[UnusualSubstance]

For topics I know a lot about I do have to edit it a little.

But it's definitely getting better and better.

Those em dashes are so damn annoying as well.

 

[Zwielicht]

I always ask it for sources on any claims it’s making. The thing about these LLMs is that they could be pulling the information from anywhere: Reddit comments, random shitty blogs, AI-spammed forums, you name it. If it’s something important such as legal or medical stuff, you especially need to make sure it’s not getting information from low-quality sources.

 

[Impulse]

I’d double check information that is important not to get wrong, like anything involving health or official documents.

And her issue was just that she needed to get an ESTA. Aren’t those really easy to get? Could have saved herself a lot of time.

Although, that article is questionably written, lol:

A Spanish influencer couple experienced a mishap that has made the internet widely known.

Good to know that more people know about the internet now. Now it can finally enter the mainstream.

 

[Rem]

And her issue was just that she needed to get an ESTA. Aren’t those really easy to get? Could have saved herself a lot of time.

It honestly would depends on where she is from, her current records & where she trying head to
Like me going into US years back took less than a few hrs for the approve email while my mate took over 2weeks before they got the approval email.
While another friend got denied lmao

But once again, chatGPT wouldn't have known her circumstances whether she needed an ESTA or not. We wouldn't know for sure unless we seen the prompt

 

[TheVigilante]

The blame game has already started

Gotta agree that it is best used as an assistant for the time being

 

[Reanimated]

No.

With Claude I do the research myself first and train it on the research. Usually comes out better than the research topics lol

Always do the leg work