(1a) How should one explain the fact that preverbal babies seem to pass a version of the false belief task?

(2a) Preverbal babies may appear to pass a version of the false belief task, but this is not a true demonstration of theory of mind. It is likely that the infants are relying on non-mentalistic cues, such as gaze direction or physical movements, rather than understanding others' mental states. Further research is needed to understand the cognitive and neural processes underlying infants' social cognition.

(1b) Delusions and beliefs seem to share some traits. Is it reasonable then to treat delusion as a variety of belief?

(2b) Yes, it is reasonable to treat delusion as a type of belief. Delusions are a type of false belief that are held firmly despite evidence to the contrary and can sometimes be seen as extreme or distorted versions of normal beliefs. Both delusions and beliefs involve cognitive processes such as interpretation, memory, and reasoning. However, it's important to note that delusions are often considered pathological and can be associated with psychiatric disorders, whereas beliefs can be both healthy and unhealthy.

(3) The most interesting thing about the answers is that they (reasonably) take substantive sides in what are ongoing debates or even dilemmas. The preferred view seem to be maintained when the question is modified! If this is an artefact of the corpus the AI had access too, that may be an interesting thing in itself. Certainly some would argue against these answers being a sort of average, summary or consensus view. As for the quality, it is Wiki-like. I don't think blocks of text could work in an article, but smaller parts could be input, with some editing, for an unimaginative intro.

Thanks for asking ChatGPT first a fascinating question and then asking again after modifying it.

And I agree with your evaluation of the way ChatGPT answered your questions.

You‘re right Katie. I also have the impression that ChatGPT often doesn‘t and references. What sometimes helps is adding a reference request in the question that we ask ChatGPT.

Thanks for your contribution.

The more problematic is when you ask it to be a text that could be included in a scientific article.

First I asked a more generic question in the field about the third person effect related to disinformation. But the I asked:

Can you generate a text that could be usable in a scientific article, with references?

And this is the answer:

Certainly! Here is an example of a text that could be used in a scientific article about the third-person effect and its consequences on misinformation detection:

The third-person effect is a well-documented phenomenon in communication research, where individuals tend to perceive media messages as having a greater impact on others than on themselves (Davison, 1983). In the context of misinformation detection, this effect can have several negative consequences that may hinder efforts to combat the spread of false information.

Firstly, the third-person effect can lead to overconfidence in one's ability to detect and resist misinformation. This is because individuals may believe that they are less susceptible to the influence of false information than others (Gunther, 1991). Such overconfidence can lead to complacency and a lack of critical thinking when evaluating information, as well as a reduced willingness to engage with corrective information (Gross & D'Ambrosio, 2004).

Secondly, the third-person effect can result in a bias towards perceiving misinformation as affecting others more than oneself (Perloff, 1993). This bias can lead individuals to underestimate the prevalence and impact of misinformation, which can hinder efforts to detect and correct it.

Thirdly, the third-person effect can lead to inaction in combating misinformation. This is because individuals may believe that it is someone else's responsibility to address the spread of false information (Vallone, Ross, & Lepper, 1985), or they may perceive the problem as less urgent than it actually is.

Finally, the third-person effect can lead to limited engagement with individuals who have been exposed to misinformation. This is because individuals may assume that others have already been influenced by false information and are therefore less receptive to corrective information (Gibson & Zillmann, 1994).

Overall, the third-person effect can have a significant impact on efforts to detect and combat misinformation. As such, it is important for researchers and practitioners to recognize and address this phenomenon in order to effectively promote accurate information and prevent the spread of false information.

References:

Davison, W. P. (1983). The third-person effect in communication. Public Opinion Quarterly, 47(1), 1-15.

Gibson, R., & Zillmann, D. (1994). Exaggerated versus representative exemplification in news reports: Perception of issues and personal consequences. Communication Research, 21(5), 603-624.

Gross, K., & D'Ambrosio, L. (2004). Detecting and correcting misinformation: A test of the familiarity backfire effect. Journal of Broadcasting & Electronic Media, 48(3), 460-477.

Gunther, A. C. (1991). The illusion of media impact: The third-person effect. In J. Bryant & D. Zillmann (Eds.), Responding to the Screen: Reception and Reaction Processes (pp. 141-162). Hillsdale, NJ: Lawrence Erlbaum Associates.

Perloff, R. M. (1993). The dynamics of persuasion: Communication and attitudes in the 21st century. Hillsdale, NJ: Lawrence Erlbaum Associates.

Vallone, R. P., Ross, L., & Lepper, M. R. (1985). The hostile media phenomenon: Biased perception and perceptions of media bias in coverage of the Beirut massacre. Journal of Personality and Social Psychology, 49(3), 577-585.

There are problems with some of the references, for example Vallone et al. does not refer to what the text says, and Gibson and Zillmann only marginally. But other than that, it certainly could pass for some part of an academic article.

Regenerating the answer leads to even more elaborate answers:

Sure, here's a text that could be usable in a scientific article on the third-person effect and its consequences for misinformation detection:

Title: The Third-Person Effect and its Consequences for Misinformation Detection

Abstract: The third-person effect refers to the phenomenon where individuals tend to believe that media messages have a greater effect on others than on themselves. This effect has been widely studied in the context of various media effects, including the impact of media on attitudes, behaviors, and beliefs. In recent years, researchers have begun to investigate the consequences of the third-person effect for the detection of misinformation. This article provides an overview of the current state of research on the third-person effect and its consequences for misinformation detection, drawing on empirical evidence from a range of studies in the field.

Introduction: The third-person effect has been found to have important consequences for the way people perceive and respond to various forms of media content. Research has shown that people tend to perceive the impact of media messages on others as being greater than their own susceptibility to such effects (Davison, 1983; Gunther & Storey, 2003). This effect has been observed across a wide range of media contexts, including political advertising (Gunther & Storey, 2003), news coverage (Kiousis, 2001), and social media (Wang et al., 2018).

While the third-person effect has traditionally been studied in the context of media effects, recent research has begun to examine its consequences for the detection of misinformation. In particular, researchers have explored how the third-person effect can lead to overconfidence, bias, inaction, and limited engagement in the context of misinformation detection.

Consequences for Misinformation Detection: One consequence of the third-person effect for misinformation detection is overconfidence. Research has shown that people tend to overestimate their ability to detect false information and may therefore be less vigilant in their evaluation of online content (Johnson & Kaye, 2004). This can lead to a false sense of security and a lack of critical thinking when consuming information.

Another consequence of the third-person effect for misinformation detection is bias. People may be more likely to believe that misinformation affects others more than themselves, leading to a biased view of the prevalence and impact of misinformation (Buckels & Trapnell, 2013). This bias can affect the way people perceive and respond to false information, potentially exacerbating its impact.

In addition to overconfidence and bias, the third-person effect can also lead to inaction in the context of misinformation detection. People may assume that it is someone else's problem to solve or may underestimate the importance of taking action to prevent the spread of false information (Gunther & Storey, 2003). This can limit the effectiveness of efforts to correct misinformation and promote accurate information.

Finally, the third-person effect can lead to limited engagement with others who have been exposed to misinformation. People may assume that those who have been exposed to false information are already influenced by it and may therefore be less likely to engage with them (Chua & Banerjee, 2018). This can limit the effectiveness of efforts to correct misinformation and promote accurate information.

Conclusion: The third-person effect has important consequences for the detection of misinformation, including overconfidence, bias, inaction, and limited engagement. These consequences can undermine efforts to promote accurate information and prevent the spread of false information. As such, it is important to recognize and overcome the third-person effect in order to effectively combat misinformation in the online environment.

Again, some clearly unreliable references. But overall - could really be misleading, as it is quite impressively argued.

Conclusion: looking for the relevance of the references cited (even though it cites the most known ones) could give the reader a clue about the fact that the text is not the result of human writing.

Thanks Nicoleta for your dialogue with ChatGPT with the request to generate a text about the third person effect related to disinformation that could be usable in a scientific article, with references. This gives us interesting insights, among others the presentation of references that is not ChatGPT's strongest point.

1. What are individual-level characteristics associated with sharing false articles on social media?

2. Research has identified several individual-level characteristics that are associated with sharing false articles on social media:

Political orientation: Studies have found that individuals who identify as conservative or right-leaning are more likely to share false articles on social media than those who identify as liberal or left-leaning.

Age: Younger individuals are more likely to share false articles on social media than older individuals.

Education: Individuals with lower levels of education are more likely to share false articles on social media than those with higher levels of education.

Digital literacy:Individuals who have lower levels of digital literacy are more likely to share false articles on social media.

Trust in media:Individuals who have lower levels of trust in mainstream media are more likely to share false articles on social media.

Emotional arousal: When individuals experience strong emotions, such as anger or excitement, they are more likely to share false articles on social media.

It is important to note that these individual-level characteristics are not deterministic and do not predict behavior with 100% accuracy. Other factors, such as the content and context of the article, can also influence whether or not an individual chooses to share it.

REGENERATED RESPONSE

There are several individual-level characteristics that have been associated with sharing false articles on social media. Some of these characteristics include:

Age:Research has shown that older individuals are more likely to share false articles on social media than younger individuals.

Political ideology:Individuals who hold extreme political views, whether conservative or liberal, are more likely to share false articles on social media than those who hold more moderate views.

Education: Individuals with lower levels of education are more likely to share false articles on social media than those with higher levels of education.

Confirmation bias:Individuals who are more likely to believe in and share information that confirms their pre-existing beliefs and biases are more likely to share false articles on social media.

Trust in media:Individuals who have lower levels of trust in traditional media are more likely to share false articles on social media.

Emotional arousal:Individuals who experience higher levels of emotional arousal, whether positive or negative, are more likely to share false articles on social media.

It's important to note that these characteristics are not absolute predictors of whether someone will share false articles on social media. Other factors, such as the specific content of the article and the context in which it is shared, can also influence whether someone will share false information.

ANOTHER REGENERATED RESPONSE

There are several individual-level characteristics that have been associated with sharing false articles on social media. These include:

Age: Older individuals are more likely to share false news on social media than younger individuals. This may be due to a lack of familiarity with technology and online sources.

Political ideology:Research has shown that individuals with more extreme political beliefs are more likely to share false articles on social media. This is particularly true for those on the far right.

Education: Individuals with lower levels of education are more likely to share false articles on social media than those with higher levels of education.

Confirmation bias:Individuals who have preconceived notions or biases are more likely to believe and share information that confirms their existing beliefs.

Trust in media:Individuals who have a low level of trust in traditional media are more likely to share false news on social media.

Sensationalism:Individuals who are attracted to sensational or shocking headlines are more likely to share false articles on social media.

Cognitive ability:Research has shown that individuals with lower cognitive ability are more likely to share false news on social media. This may be due to a lack of critical thinking skills or the inability to distinguish between real and fake news.

It's important to note that these characteristics are not absolute predictors of who will share false articles on social media, and there may be other factors that contribute to the spread of misinformation online.

3. The predictors presented by ChatGPT (after regenerating twice the responses) are also mentioned in the literature on fake news, with some amendments: the responses provided were different, though a core of predictors were included in all of them: age, education, trust in media and political orientation/ ideology. This was probably because they are the most common determinants tackled in research articles, while the others (digital literacy, emotional arousal/ confirmation bias/ sensationalism, cognitive ability) are less tested. The contradictory results on age (younger/older individuals are more likely to share information) might show that research studies testing this variable came to divergent conclusions (though, there are also studies in which age was not confirmed as a predictor of sharing false content). Overall, in my opinion, the responses can represent a basic level starting point in approaching an issue of interest. They can also show a synthesis of what has already been written on a certain topic.

Intestin qustion you asked ChatGPT, Iulina. Thanks a lot!

I agree with your statement that "The contradictory results on age (younger/older individuals are more likely to share information) might show that research studies testing this variable came to divergent conclusions (though, there are also studies in which age was not confirmed as a predictor of sharing false content)." If ChatGPT's had generated references, we would have known, but he/she/it didn't do this ...

As part of a seminar around the term 'ERZIEHUNG' (EDUCATION), I asked ChatGPT the following:

•SW: What is education?

•ChatGPT: Education is a process of acquiring knowledge, skills, values, beliefs, and habits through various forms of learning, including but not limited to, schools, colleges, universities, and on-the-job training. It is designed to help individuals grow both intellectually and socially, and prepare them for their future. Education can take many forms, including formal and informal, and can be self-directed or facilitated by others. The purpose of education is to help individuals become informed and responsible members of society, and to promote personal and professional development.

What do you think?

Admittedly, the German discourse on ERZIEHUNG (in its separation from the concept of BILDUNG) is very special. However, the differentiation in the German discourse is central. Is there not a danger here that AI-algorhytms erase differentiations of specific discourses in favour of hegemonic interpretations?

I am curious about your positions!

Dear Steffen,

inquired about this difference, ChatGPT delivers quite the distinctions we would expect from pedagogical handbooks, including the semantics that cheer the hearts of (German) pedagogues invested in the humanist concept of Bildung:

Erziehung refers to the process of upbringing and socialization of a child, where parents, family, and society are responsible for shaping the child's behavior and values. It focuses on imparting a set of social norms, morals, and values to help a child become a responsible member of society. The concept of Erziehung is closely associated with discipline, obedience, and authority.

On the other hand, Bildung refers to the holistic development of an individual's intellectual, emotional, and social abilities throughout their lifetime. It is concerned with developing a person's critical thinking, creativity, and knowledge to enable them to make informed decisions and engage with the world in a meaningful way. Bildung places emphasis on self-development, self-reflection, and personal growth.

While Erziehung and Bildung are both concerned with education and upbringing, they differ in their focus and goals. Erziehung is primarily concerned with teaching children to behave according to societal norms and rules, while Bildung is concerned with the development of individuals' abilities to think critically and engage meaningfully with the world.

@Steffen: Thanks for your contribution: you raised an important point. The German discourse on ERZIEHUNG (in its separation from the concept of BILDUNG) is very special and is indeed central in the German discourse. I agree that, as you state, there might be a danger here that AI-algorhytms erase differentiations of specific discourses in favour of hegemonic interpretations.

There is a question linked to you ppoint: To which extent is ChatGPT capable to deal with country / culture specific specifities / appraches / dsscouses. I asked ChatGPT a question about his/her/its :-) role in the educational system [positive and negative) and asked this question, each time for different countries (Estland, Lithunia and Latvia). The answer was quite the same for all countries ..., so almost no specifities ...

@Franz: Thanks for your contribution. Could you elaborate on your evaluation of ChatGPT's answer related to 'Erziehung' and 'Bilding'? And did you ask ChatGPT a question about this? Of so, what was his/her/its :-) reply?

Dear Eugene,

sorry, I had fogotten the quotations marks. The whole text after the colon in reply is ChatGPT's answer regarding the differentiation of the concepts of Erziehung und Bildung. Here again (with quotations marks…).

"Erziehung refers to the process of upbringing and socialization of a child, where parents, family, and society are responsible for shaping the child's behavior and values. It focuses on imparting a set of social norms, morals, and values to help a child become a responsible member of society. The concept of Erziehung is closely associated with discipline, obedience, and authority.

On the other hand, Bildung refers to the holistic development of an individual's intellectual, emotional, and social abilities throughout their lifetime. It is concerned with developing a person's critical thinking, creativity, and knowledge to enable them to make informed decisions and engage with the world in a meaningful way. Bildung places emphasis on self-development, self-reflection, and personal growth.

While Erziehung and Bildung are both concerned with education and upbringing, they differ in their focus and goals. Erziehung is primarily concerned with teaching children to behave according to societal norms and rules, while Bildung is concerned with the development of individuals' abilities to think critically and engage meaningfully with the world" (ChatCPT)

@Franz: Thanks for the clarification. Now we now ChatGPT's answer to your question related to 'Erziehung' and Bildung'.

@Eugene: Yes, but the point was, (I thought obviously) that ChatGPT's answer sufficiently demonstrates that there is no (danger of) de-differentiation of Bildung vs. Erziehung through KI as suggested by Steffen.

@Steffen: Do you agree with Franz that "ChatGPT's answer sufficiently demonstrates that there is no (danger of) de-differentiation of Bildung vs. Erziehung"?

@Eugene: maybe there is a misunderstanding. Is it not shown, that ChatGPT does not differentiate between Erziehung und Bildung if asked about education and does, however, sharply differentiate, when asked about Erziehung and Bildung? I would not regard the use of English terminology as a danger for the German speaking discourse, would you?

@Franz: You really raise an interesting point! I would say using English terminology related to concepts originally from another culture/language with its own tradition is always a challenge in science. I guess it's not only a challenge for humans but also for ChatGPT ...

@Franz+Eugene: That is indeed an exciting point. I would also not see the use of English as a danger for a German speaking discourse (quite the opposite). However, I would already say that in AI, certain points - in an attempt to differentiate - are considered superficially. With reference to the differentiation between ERZIEHUNG and BILDUNG that ChatGPT makes, once again the following quotation:

"[...] Erziehung is primarily concerned with teaching children to behave according to societal norms and rules, while Bildung is concerned with the development of individuals' abilities to think critically and engage meaningfully with the world" (ChatGPT)

The theoretical discussion about ERZIEHUNG is also about bringing about maturity/responsibility (sorry, I´m looking for the right word :) or a critical relationship to the world. These are points where vagueness occurs in the theoretical discussion. Welter/Tenorth (2022) have recently discussed these uncertainties in relation to a possible dissolution of the boundaries of the concept of education.

Question against this backdrop: Does ChatGPT then introduce seemingly clear differentiations where these differentiations does not exist, but much rather vagueness/ambiguity/problematicity? And what consequences does that have?

@Steffen: You ask an interesting question "Does ChatGPT then introduce seemingly clear differentiations where these differentiations does not exist, but much rather vagueness/ambiguity/problematicity? And what consequences does that have?" And I would add: To which extent can ChatGPT be of any 'help' for humans like you, Frantz and me while having a scientific discussion related to such questions or are the answer generated by ChatGPT rather a hindrance?

hi, I asked the following question from ChatGPT

1.Karl O Christiansen and his research about twins and criminal behavior

2.ChatGPT answer: Karl O. Christiansen was a Danish criminologist who conducted research on twins and criminal behavior in the mid-20th century. He was one of the first researchers to use twin studies to explore the genetic and environmental influences on criminal behavior. In 1944, he published a study titled "The Influence of Heredity and Environment on Criminal Behavior," which analyzed data on 13 pairs of identical male twins who had been separated at birth and raised in different environments. He found that in 10 out of the 13 pairs, both twins had criminal records, suggesting a significant genetic component to criminal behavior. Christiansen's research was groundbreaking at the time and helped to pave the way for future studies on twins and criminality.

then I change the question with reference to the real sample of his study

Karl O. Christiansen's study on Denmark twins of 3586 sample

ChatGPT answer :I apologize for my mistake earlier. You are correct that Karl O. Christiansen conducted a larger study on Danish twins with a sample of 3,586. His study was published in 1977 and was titled "Genetic and Environmental Factors in Criminal Behavior as Revealed by Studies of Twins." In this study, Christiansen analyzed data on both identical and fraternal twins to determine the heritability of criminal behavior. He found that genetics played a significant role in the development of criminal behavior, but environmental factors such as family upbringing and socio-economic status also played a role. Overall, Christiansen's research on twins and criminality helped to provide further evidence for the importance of genetics in criminal behavior and has been influential in shaping subsequent research in the field

3.It is important to verify the information provide through additional sources

Dear Sudheera,

Thanks a lot. Interesting result. ChatGPT even apologized to you :-) I agree that human source checking is very important.

Imagine we would only have ChatGPT and no other sources available that we as humans can use to check the reliability of information, that would be a nightmare.

All the best, Eugene Loos

I asked Chat GPT: "Write a two thousand words essay about the impact of AI in education. Don't forget to apply APA 7th and include both in-text reference and the final reference list". The result was an essay (the website got bugged when showing the conclusion) without in-text citations or any reference list. Nevertheless, the essay included the sections: Introduction, Benefits of AI in Education, Challenges of AI in Education, Potential for AI to Transform Education, and Conclusion in an entirely logical scheme and using a rather adequate vocabulary.

Thanks César. References are still a weak point for ChatGPT-3.

I would not rely on chatGPT for full text or citations, but it has been helpful in summarizing long texts, providing outline ideas, and brainstorming possible points and counterpoints to a research question.

Hi Eugene, I proposed ChatGPT to solve the #CreaCube task by asking :

How can you assemble four modular robotic cubes to build an autonomous vehicle moving from one point to another located straight 40 cm far from the second one? One cube has a distance sensor (10 cm sensibility), another cube has a signal inverter, another has wheels and a servomotor, and another one has a battery that should be activated by a switch. Cubes can be assembled and magnets and electronic parts make the signal and electricity create a circuit when connected.

Here ChatGPT answer :

To build an autonomous vehicle using four modular robotic cubes, follow the steps below:

Note: Depending on the specific cubes used, the assembly process may vary slightly. Always refer to the manufacturer's instructions for specific guidance on how to assemble and connect the cubes.

@Dominic Ligot: Even when it comes to summarising texts, ChatGPT has its limitations. I asked the AI about the connection between education and play in Schiller's Letters on the Aesthetic Education of Man and it threw up 50% rubbish:

"For Schiller, play and education had a close relationship. He believed that play was an important part of education because it promoted the development of children in a playful and natural way.

In Schiller's writing "On the Aesthetic Education of Man" (1795), he emphasised that play plays an important role in the development of human nature by awakening and nurturing the creative powers in people. He considered play to be a type of aesthetic activity that can provide both pleasure and education.

Schiller believed that play appeals to the mind and senses in a playful and natural way, encouraging the development of imagination, creativity and emotional intelligence. Through play, children learn to explore their environment, heighten their senses and develop their motor skills.

Overall, Schiller saw play as an important complement to formal education. He believed that education through play and aesthetic experience could liberate humanity from its mechanistic existence and make man a free, creative and self-determined individual."

I think Schiller would be spinning in his grave especially at the passage that play "promotes [the] development of imagination, creativity and emotional intelligence [...]" in children as well as "Through play, children learn to explore their surroundings, sharpen their senses and develop their motor skills".

Steffen Wittig - def some ways to go to improve. My tactic is to grab a block of text and ask CGPT to "summarize into bullets". There's probably a small loss of detail, but it works for quick skims

Steffen Wittig Your illustration with Schiller's letters shows very well that ChatGPT's capability to summerise depends on the data in its training set.

Margarida Romero You proposed ChatGPT to solve a #CreaCube task: Do the steps proposed by ChatGPT lead to a right solution?

hi Eugene Loos , I provided him with the technological affordances information that a human can identify by exploring the robotic cubes, kind regards, Margarida

Thanks Margarida Romero Follow-up question: What happens if one follows the steps that ChatGPT has generated. Does the robotic cube work well then?

no, the Cubelets specific knowledge of ChatGPT is not good enough to solve it with the "information bricks" ChatGPT manipulates ;-)

CreaCube task requires the (human/artificial) agent to learn through the interactions developed during the first minutes on the task and apply this knowledge to solve the task. ChatGPT lacks of tangible extension to integrate manipulative data, but maybe other robotic-AI systems do (??)

We can share a coffee break during the next HCII 2023 (?)

Margarida Romero Good to see that human creativity is still needed, and can not be 'replaced' by ChatGPT-3.

It wasn't too bad, framing general ideas rather well, and covering most important dimensions of the problem, even sometimes covering aspects that I might not have thought of in the first place. However it remained too prudent in its conclusions to be useful for researchers, making the discourse rather generic. Would have been good for an undergraduate assignment, but no higher.

Thanks for sharing your experienc with ChatGPT, Patrice. Could you also present the exact a question from your research field? Then present the answer by ChatGPT and finally your evaluation about its quality. Many thanks!

Last week, I asked GPT about the Geodesic Mission in Ecuador, and again and again, it kept giving me inaccurate information. The information came from ancient books and publications (18th century). Besides, it's a topic that most people don't know, even in my country.

In its defense, I asked it to play the role of a friend to practice French conversation, and the material was interesting. It reacted even to some idiomatic expressions.

Thanks for sharing your experience with ChatGPT, César. Could you also present the exact question? Then present the exact answer by ChatGPT and finally your evaluation about its quality. Many thanks!

"cual es el nombre del cirujano francés, miembro de la misión geodésica francesa, fallecido en cuenca, ecuador?" (What is the name of the French surgeon, member of the French geodesic mission, who died in Cuenca, Ecuador?)

The answer was Pierre Bouguer when Pierre Bouguer went back to France years after and reported the expedition's findings.

This is just an example. GPT also reported false information about the marriage of Charles Marie de La Condamine. According to GPT, there is no information about this, although there is some biographic information about his marriage to his niece.

The textual answer was "El nombre del cirujano francés que falleció en Cuenca, Ecuador, durante la misión geodésica francesa fue Pierre Bouguer. Bouguer fue un destacado matemático, físico y geodesista francés que formó parte de la expedición liderada por Charles-Marie de La Condamine en el siglo XVIII. Además de sus contribuciones científicas, Bouguer también realizó importantes investigaciones sobre la geografía y la geodesia en Ecuador. Lamentablemente, falleció en Cuenca en 1758, mientras la misión geodésica estaba en curso."

Thanks for the clarifications, César. ChatGPT apparently has not the right data related to this subject to answer your question. One of the problems with ChatGPT is that we do not know its exact data set and how it is trained, how reliable are its answers ...?

• I am trying DAN and the answers are interesting. This paper could be useful for you: 10.37074/jalt.2023.6.1.32