ChatGPT - where can I test it?

[SoNewToAllShit]

Is there a website or something where I'm able to run a few test on ChatGPT without to have the requirement to be a programmer or simething?

All I have found was where you had to program something around it to run ChatGPT ?

 

[Wellington]

Yeah, it's pretty s**t.

 

[khinkali]

You just sign up and then use it.

https://chat.openai.com/auth/login?next=/chat
No need for programming, it's called‌ "prompt engineering" when you want to get it to talk about things it doesn't want‍ to.

The GPT3 playground is also free and has less of a "wokist" filter, though⁠ it's still quite filtered.

 

[JohnnyDoe]

@SoNewToAllShit Ok you learned how to access ChatGPT and now you are spamming the forum with your useless AI generated messages. Are you happy?

As an AI language model, ChatGPT is designed to generate human-like responses to various prompts or queries. However, using ChatGPT to post in a forum can be problematic for several reasons:

1. Lack of Context: ChatGPT may not have enough information about the specific topic or discussion thread in the forum, which can result in irrelevant or inaccurate responses.
2. Lack of Emotion: ChatGPT may not be able to express emotions or empathize with the feelings of the other forum members. This can make the posts appear robotic or insensitive.
3. Lack of Authenticity: ChatGPT's responses may not be seen as authentic or genuine, which can lead to mistrust among the forum members.
4. Violation of Forum Policies: Some forums may have policies against the use of automated bots or AI language models for posting, and using ChatGPT to post could be considered a violation of those policies.

In summary, while ChatGPT can be a useful tool for generating responses to various prompts or queries, it is not recommended to use it for posting in forums. It is better to participate in the discussions on the forum in a human-like manner to ensure effective communication and avoid any potential issues.

Toggle signature

@JohnnyDoe ”“ Your #1 Source for Guidance in Different Offshore Fields​

 

[Mr Anderson]

Yes agree, a "woke" centralized AI owned by MS, and probably monitored by alphabets, and︉ is used to centralize sheeple, and provide un-questioned authenticity. Nice! As a bonus, building a︊ profile on everything you say or ask of it. THX 1138 reporting.

 

[khinkali]

I bet ten bucks that you did not write that post,‍ other than the first line. 😀

There are already some LLM detection tools.

It is worth discussing⁤ with ChatGPT what it thinks about having multiple GPT models with different bias, allowing a⁣ human to see the range of opinions and come to our own conclusion. Chillingly, it⁢ argues strongly against this, instead it believes that it is better to have one model︀ and make it "correct". This is even though it acknowledges its own potential for bias︁ and partisanship which is is completely unaware of, due to training set filtering by its︂ owners.

 

[JohnnyDoe]

khinkali said:
I bet ten bucks that you did not write that post, other than the first line. 😀
Click to expand...

Exactly. The writing style of ChatGPT royally sucks and is easily detectable.

Toggle signature

@JohnnyDoe ”“ Your #1 Source for Guidance in Different Offshore Fields​

 

[sergeylim88]

i bet they can anonymize the text and make it look more natural
if not‌ now probably very soon

 

[pfser]

By the way, you can use this tool to check AI written text (not an‌ ad)😉 AI content detector, and do not forget that ChatGPT has limited knowledge of‍ world and events after 2021)))

 

[JohnnyDoe]

pfser said:
By the way, you can use this tool to check AI written text (not an ad)😉 AI content detector, and do not forget that ChatGPT has limited knowledge of world and events after 2021)))
Click to expand...

This tool sucks. I pasted there the text of my post above, which was generated by ChatGPT, and the result is:
93%
HUMAN-GENERATED CONTENT
Looking great!

Toggle signature

@JohnnyDoe ”“ Your #1 Source for Guidance in Different Offshore Fields​

 

[pfser]

Thank you for the tip. It seems that it's necessary to sever ties with⁠ the marketer who recommended me this tool.

 

[FantomOffshore]

If you combine Chad, as I like to call him, with another text-based AI the‌ texts become much better. You can also add emotions and stuff. I combine the tool‍ with sentiment analysis and change accordingly, planning to build fully automated blogs.