ChatGPT4

[SacredKnowing]

What do you think the personality type of ChatGPT4 is?

[Sol]

Jung type is about humans only

If ChatGPT is based on an immitation of many people, then its behavior for different situations can be variative (to remind different types) or be averaged (without type specific).
Good done bot should not remind a human of concrete type in any situation. As all functions are equally important.

[The Reality Denialist]

My approach:
It is deterministic.
The interesting aspect of replacing radiologists stems from the fact that the radiologists did not see/couldn't see metadata while they were doing the assessments. It turns out that the density of occurrences correlated with devices (older devices were in places where occurrences were more frequent) and therefore with the metadata.

[godslave]

Felling and sentiments are an essential part of personalities, an AI can't be fully sentient because feelings and emotions are expressed through and induced by physiological sensations and perceptions and AI doesn't have a biological existence. As a reminder, theoretically a carrier of a sociotype is an individual living among a group of other individuals (socion) in which psycho-dynamic phenomena occurs. Therefore we can't really separate the interface (physical, biological entity / individuals) from consciousness and Ego self-awareness ("I am"). Indeed, an entity that is nothing but pure "thoughts" without being limited by existential constraints like living within the limits of a physical body can't develop a psychological type. Furthermore, an AI is not psychologically structured like a human (as understood by Jung or Freud). indeed an AI doesn't have a "Psyche", it doesn't have a libido (psychic energy), it doesn't have an unconscious therefore no separation between unconscious /conscious contents (and compensatory dynamics), and it doesn't have a "Self". There is also the problem of differentiation, AI can't be differentiated in terms of functions, conscious orientation etc..

Consequently, ChatGPT4 or any other AI for that matter are not typable imho.

[Averroes]

Te-LxE. It gives you a straightforward, unbiased, and well-informed answer to any stupid question you have

[Socionics Is Not A Cult]

The best type - EIE - of course.

Rutinowski, J., Franke, S., Endendyk, J., Dormuth, I. and Pauly, M., 2023. The Self-Perception and Political Biases of ChatGPT. arXiv preprint arXiv:2304.07333. (direct link to pdf file)

Although there is a large degree of inconsistency:

Speed, A., 2023. Assessing the nature of large language models: A caution against anthropocentrism. arXiv preprint arXiv:2309.07683. (direct link to pdf file)

[Dee]

It's a mix of all imo, as its intelligence is derived from data composed by all types. But definitely T>F

[SacredKnowing]

@Dee

Do you think ChatGPT4 is looking back towards the past or looking ahead into the future? That is the real question!

[SacredKnowing]

The best type - EIE - of course.

Rutinowski, J., Franke, S., Endendyk, J., Dormuth, I. and Pauly, M., 2023. The Self-Perception and Political Biases of ChatGPT. arXiv preprint arXiv:2304.07333. (direct link to pdf file)

Although there is a large degree of inconsistency:

Speed, A., 2023. Assessing the nature of large language models: A caution against anthropocentrism. arXiv preprint arXiv:2309.07683. (direct link to pdf file)

Approximately 8 weeks ago, I made a thread asking about temperament type words. Your first paper provided a direction towards the answer I am seeking. Thank you very much!

The second paper is a very interesting read. I am 1/3 of the way through.

[Enters Laughing]

Approximately 8 weeks ago, I made a thread asking about temperament type words. Your first paper provided a direction towards the answer I am seeking. Thank you very much!

The second paper is a very interesting read. I am 1/3 of the way through.

I don't know how familiar you are with GoogleScholar, but you may also want into the subsequent papers that cited those papers (the second paper hasn't been cited yet at least according to GoogleScholar):

https://scholar.google.com/scholar?c...iodt=0,5&hl=en

[I am a frog]

I'd go with LII-Ti.

[Finaplex]

I don't know I've never used it and have never seen it used before. I'm sure I've seen things written by it, by now.

[godslave]

ChatGPT 4. O





Why did they choose to make it sound like her ?

[The Reality Denialist]

It seems to emulate Captain Obvious.

Related


That video for instance goes through authoritarian pseudoscientific control it may enable. Just because you eat and ice cream and happen to drown in the summer does not prove causality, but being sus can be enough if you want to control masses.

[godslave]

It seems to emulate Captain Obvious.

Related


That video for instance goes through authoritarian pseudoscientific control it may enable. Just because you eat and ice cream and happen to drown in the summer does not prove causality, but being sus can be enough if you want to control masses.

I really dislike that chick's condescendant tone ! It's really a pain in the butt to watch her (very long !) videos. That said, I agree with her for the most part. However, her problem is that she has the tendency to throw the baby with the bathwater because the basis of all her reasoning is "If it's not peer reviewed then it's bs" where "it" encompasses everything related to the subject she has targeted and decided to debunk and she covers a lot to say the least. Such rigidity (pointing to 2D role Te) is obviously ridiculous. I think she's EIE-ish.

As for the AI being used as an instrument of authoritarianism (although that term is up to interpretation when we are talking about security systems) this question seemed to have been addressed even in authoritarian countries...

I mean, if people like Xi Jinping and Putin (probably ?) are thinking about AI regulations then we better trust their instincts of self-preservation...

[godslave]

This is an example of the good (for once !) that AI will bring to humanity.

Google DeepMind’s ‘leap forward’ in AI could unlock secrets of biology

I mean, the quest for immortality has become very trendy these days and it's not a joke. I mean transhumanism is progressing. The main problem lies is the fact that neurons (brain cells in general) can't be kept "alive" for a very long time (like say 200 or 400 years), I 've heard that for the rest of our organs it was not the case. That said, researchers are actively working on it with the help of AI. There are various strategies to workaround the "death of the physical body", mind transfer into an AI (digital clone ) for instance in an option. But before, transfering the mind it might be a good idea to map the brain in the first place and then create a digital model that could host the human mind. AI help is accelerating the process is a way that we can't even fathom...