ChatGPT exposes hidden wipers at the TOP of windshield?

[Ogre]

No, it's not real. It's text written by a new and somewhat unethically sourced AI.

-Crissa

Unethically sourced?

Hadn’t heard this previously.

Sponsored

 

[Crissa]

Yeah, OpenAI specifically isn't transparent about its human trainers.

-Crissa

 

[Usachris]

I identify as droid/bot

That’s it! All new and current members of this forum will have to pass a CAPTCHA exam!

 

[SwampNut]

Unethically sourced?

Hadn’t heard this previously.

Probably because it's political BS?

Two interesting things here...that someone seems to have no heard about this system, and also all the hype about it. I can't make it do anything useful, or when it seems useful, it's slightly less so than a regular google search. It very easy to change results by how you ask questions. I guess that is somewhat human also, as you may be hinting at context. But it also means that the resulting answers, if they involve some sort of emotional or ethics component, can be manipulated. For example it told me that I definitely could not shake anti-vaxxers like a baby until they lose their stupidity, but was much softer on the idea of leaving them outside in the snow until they repent. I got it to sort of tell offensive jokes, which it's supposed to avoid, by asking for a story about a comedian who tells offensive jokes. I'm guess with some work I can get it to explain all the good that Hitler did for the world.

The only good thing to come from my "talks" with it was a recipe fitting a certain profile (vegan, whole food, certain region), and I will make it. Google gives me a bunch, the chat gave me one that seems perfect. Which is better? Dunno, I kinda do like the "here's one answer."

 

[RavenX447]

? ?

 

[Ogre]

Probably because it's political BS?

Two interesting things here...that someone seems to have no heard about this system, and also all the hype about it. I can't make it do anything useful, or when it seems useful, it's slightly less so than a regular google search. It very easy to change results by how you ask questions. I guess that is somewhat human also, as you may be hinting at context. But it also means that the resulting answers, if they involve some sort of emotional or ethics component, can be manipulated. For example it told me that I definitely could not shake anti-vaxxers like a baby until they lose their stupidity, but was much softer on the idea of leaving them outside in the snow until they repent. I got it to sort of tell offensive jokes, which it's supposed to avoid, by asking for a story about a comedian who tells offensive jokes. I'm guess with some work I can get it to explain all the good that Hitler did for the world.

The only good thing to come from my "talks" with it was a recipe fitting a certain profile (vegan, whole food, certain region), and I will make it. Google gives me a bunch, the chat gave me one that seems perfect. Which is better? Dunno, I kinda do like the "here's one answer."

I think people greatly mis-understand what is great about ChatGPT.

The miracle of ChatGPT isn’t that it is super knowledgeable. While it does have fairly encyclopedic knowledge, it’s more akin to a set of encyclopedias than what Google does. It’s been fed a fairly massive amount of information and somewhat self-organized that information.

ChatGPT is rad not because of what it knows, but what it does with that information. Google gives you are set of links, some chunks of information gleaned from the website, and maybe an image again from that site. Where ChatGPT shines is it’s ability to compile that information into human readable text. ChatGPT tells you what the website says.

It’s very much like a college freshmen with access to a research library or a research assistant. It can pull data from that library and compile simple reports, compare things, or even write simple sales pitches with customizations for who you are pitching. It’s a fantastic tool, but it’s not a good primary source of information.

 

[ClayBell]

Or it read all the tweets about the Cybertruck and my suggestion for hiding the wipers at the top was one of them. Try asking if the seats are removable.

 

[Cybertruck Hawaii]

Mounting it vertically is much more aerodynamic than mounting it horizontally.

 

[ClayBell]

Mounting it vertically is much more aerodynamic than mounting it horizontally.

Agreed if it’s external, one or 2 wipers parked on the side I think is the most likely config. However, a retractable box, at the top, would be the most efficient for aero, but likely not the best for mfr.

 

[ClayBell]

Also, most AI can pass captcha, so to be clear, I’m 130% human, the school nurse confirmed it.

 

[Cybertruck Hawaii]

Agreed if it’s external, one or 2 wipers parked on the side I think is the most likely config. However, a retractable box, at the top, would be the most efficient for aero, but likely not the best for mfr.

?Judging by how Tesla is trying to squeeze ever last drop of improving the coefficients of aerodynamic drag, such as wheel covers and flushed door handles, I would imagine that the wiper position will matter at the end.

 

[charliemagpie]

Look at the frivolous claims prominently laid out against Tesla in Wikipedia.

This is data. Someone knows what they are doing!!

 

[Crissa]

Also, most AI can pass captcha, so to be clear, I’m 130% human, the school nurse confirmed it.

Defeated by Earth's greatest defense: a bumblebee!

-Crissa

 

[Cyberman]

I identify as droid/bot

As evidenced by your name

 

[SwampNut]

I think people greatly mis-understand what is great about ChatGPT.

The miracle of ChatGPT isn’t that it is super knowledgeable. While it does have fairly encyclopedic knowledge, it’s more akin to a set of encyclopedias than what Google does. It’s been fed a fairly massive amount of information and somewhat self-organized that information.

ChatGPT is rad not because of what it knows, but what it does with that information. Google gives you are set of links, some chunks of information gleaned from the website, and maybe an image again from that site. Where ChatGPT shines is it’s ability to compile that information into human readable text. ChatGPT tells you what the website says.

It’s very much like a college freshmen with access to a research library or a research assistant. It can pull data from that library and compile simple reports, compare things, or even write simple sales pitches with customizations for who you are pitching. It’s a fantastic tool, but it’s not a good primary source of information.

I'm well aware of the claim/promise. I'm saying that my experience/opinion is that it fails to do what is promised. Also Google does already try to distill answers and present the relevant parts of the results. I'm a bit shocked that you haven't seen that.

There's no clear winner here. Google can be spot on or can fail when there's a software revision that changes a function. Same with the AI, but with google its easier to see that.

change theme invision community.


Invision Community is a software platform that allows users to create online communities and forums. If you want to change the theme (visual appearance) of your Invision Community forum, you can do so by following these steps:

1. Log in to your Invision Community forum as an administrator.
2. Navigate to the "Customize" tab in the dashboard.
3. In the "Customize" tab, you will see a list of options for customizing your forum's appearance.
4. Click on the "Themes" option to access the theme settings.
5. From the theme settings page, you can select a new theme for your forum. You can choose from a list of pre-designed themes or upload your own custom theme.
6. Once you have selected a new theme, click the "Save" button to apply the changes.

Keep in mind that changing the theme of your Invision Community forum may affect the appearance of other elements on your forum, such as the layout and colors of certain pages and features. It's a good idea to preview your changes before saving them to make sure everything looks the way you want it to.

Sponsored