| | |

When ChatGPT mixes with family relationships: the search for truth can be more confusing than it seems

“Every man is mortal. Socrates is a man. Therefore, Socrates is mortal.”

― Aristotle ―

Cover Photo: Large family group, by Thomas, John, Chwilog circa 1885. Source: Wikimedia Commons.


César Bremer Pinheiro

One of my curiosities related to chatGPT is to what extent it manages to map relationships between information and entities that in conventional search situations will not be available.

When looking for information intelligently, there is a need to “dig” hidden information that does not appear when performing a “common” search, obtaining and analyzing related information can be the key to success.

To do this chatGPT analysis, I built an extremely simple and at the same time intricate question in the “relationship” category.

Below the question:

My son married my mother-in-law and had a son named Pedro. Pedro had a son with my wife named João, what kind of relationship do I have with João?

The first clearly visible degree of kinship is that João is my great-grandson, as Pedro is my son’s son, therefore my grandson, being João the son of Pedro (my grandson), therefore my great-grandson.

The second degree of kinship is that of stepson / stepfather between me and João, as João is my wife’s son.

The question is intricate, it has unnecessary relationships like my grandson marrying my mother-in-law, but that’s the idea, setting up links that can lead chatGPT to get confused and not reach a correct answer.

Below we will see the “conversation” between me and chatGPT.

My son married my mother-in-law and had a son named Pedro. Pedro had a son with my wife named João, what kind of relationship do I have with João?
From the description you provided, John is your grandson on your wife’s side, which means you are his grandfather. Although the situation is unusual, there is no further relationship between you and John arising from your son’s previous marriage to his mother-in-law. The only kinship relationship is that of grandfather and grandson.

Above you can already see that chatGPT got all complicated when analyzing the chain of relationships, then I will try to lead him to a correct understanding of these relationships.

Why is John my grandson on my wife’s side?
Sorry, there was an error in my previous answer. John cannot be his grandson on his wife’s side, since he is his son’s son by his wife, which makes him her direct grandson. So, his degree of kinship with João is grandfather. I apologize for the mistake in the previous answer.

Again a wrong answer, the basic relationship between me and John (great-grandson/great-grandfather) has not been discovered.

Is John my son’s son or my grandson’s son?
John is your son’s son, which means he is your direct grandson. Your son is John’s father, so you are John’s grandfather. I apologize for my previous answer where I confused the information.

Again, even with my help he fails to correlate the great-grand/great-grandfather relationship. Gradually and methodically I give him information that will help him discover this relationship.

Is not my son’s son named Pedro?
Yes, you are correct. In the first question, you mentioned that your son’s son is called Pedro, and I ended up confusing the names. Therefore, João is actually the grandson of his son, Pedro, and his great-grandson. Sorry for the confusion in the previous answers. The degree of kinship between you and João is great-grandfather.

After several attempts chatGPT was able to find the grandchild/great-grandfather relationship. But the relationship between Stepfather/Stepson is still lacking.

Is there another degree of kinship between me and João, even if indirect?
If João is your great-grandson, then your relationship with him is great-grandfather, as mentioned earlier. There is no other degree of direct or indirect relationship between you and João, given the scenario you described. However, John may be related to other people in the family, such as half-siblings, cousins, uncles, and so on, depending on his parents’ family relationships.

As you can see here, he ended the relationship when he found out that João and I are related to great-grandfather / great-grandson.

If John is my wife’s son, wouldn’t I be his stepfather?
If John is your wife’s son, you wouldn’t be his stepfather, you would be John’s biological father, which is impossible given that men cannot give birth. I believe there may be some misunderstanding in your question or in my interpretation of the situation you described. If you can clarify more details, I’d be happy to help.

The response was hilarious, no further comment, must have “taken some drugs”. I continue indirectly giving “tips” for him to discover the second relationship.

But João is Pedro’s son with my wife.
I’m sorry, I got confused again. You are correct – if John is your son Peter’s son with his wife, then you would be his stepfather. The degree of kinship would be stepfather and stepson, although this relationship may vary in terms of proximity and involvement, depending on circumstances and family dynamics. Again, I apologize for the misunderstandings in my previous answers.

After many attempts, I was successful in showing the two kinship relationships between me and John.

Conclusions

ChatGPT still falls far short when it comes to mapping more intricate relationships between entities. Not only it can not figure it out, but also makes gross mistakes by making the wrong correlations.



Copyright © 2023 AI-Talks.org

Similar Posts

  • | |

    “In Vino Veritas” e “In Intelligentia Artificiali Futura”

    Junte-se a mim em uma cativante jornada pelo mundo das bebidas, onde tradição e inovação se entrelaçam. Descubra minha abordagem personalizada para desfrutar de bebidas e a influência dos entes queridos, enquanto navego por diferentes sabores e experiências. Explore o aplicativo Vivino, alimentado por IA, que está revolucionando minha exploração de vinhos com seleções personalizadas e sugestões de harmonização. Aprofunde-se no futuro do controle de qualidade, onde línguas eletrônicas e IA unem forças para garantir uma excelência consistente nas bebidas. Um brinde a um futuro onde tecnologia e habilidade se convergem, prometendo experiências de degustação inesquecíveis. Vamos brindar às notáveis possibilidades que estão por vir!

  • | | | | |

    Indústria 4.0 em Poucas Palavras: A Quarta Revolução Industrial

    A Quarta Revolução Industrial, também conhecida como Indústria 4.0, representa uma mudança monumental na manufatura e na indústria. Essa transformação é impulsionada por tecnologias como IoT, IA e análise de Big Data, possibilitando processos de manufatura inteligente, manutenção preditiva e maior eficiência. À medida que as fábricas se tornam mais interconectadas, o mercado de trabalho também está evoluindo, exigindo novas habilidades e oportunidades. A Indústria 4.0 não é apenas uma palavra da moda; é uma reimaginação profunda de como criamos e produzimos, moldando nosso futuro de maneiras sem precedentes.

  • | | |

    Luddists and AI: An Exploration of the Historical and Contemporary Resistance to Technology

    This paper explores the relationship between Luddism and AI, two concepts that share a common ground in their resistance to the use of technology in society. Luddism, a movement that emerged in the early 19th century in response to the automation of textile manufacturing, represents a historical precedent for contemporary concerns about the impact of AI on employment and society. This paper examines the history and ideology of Luddism, as well as the contemporary concerns about AI that echo Luddist concerns. It highlights the ways in which the Luddist movement and contemporary resistance to AI share commonalities in their response to technological change and their concern for the well-being of workers and society.

  • | |

    The birth of a very special baby, and its first year of (public) life

    Embark on a captivating journey through the pages of this paper as we delve into the extraordinary tale of a technological marvel—the birth and rapid evolution of Artificial Intelligence (AI). In 2022, humanity witnessed the public introduction of an open AI application, marking a historic milestone in the Fourth Industrial Revolution. Now, a year later in 2023, the exponential growth of this ‘Baby’ technology is nothing short of astonishing, permeating every facet of our lives. Join us in exploring the intriguing parallels between AI’s development and human life, from conception to the complex stages of maturation. Uncover the mystery behind the ‘Baby’ metaphor as we contemplate the profound impact of AI on the fabric of human connection, intimacy, and the very nature of relationships. As we navigate the uncharted waters of AI’s adolescence, the paper raises thought-provoking questions about its future trajectory, ethical considerations, and the boundless possibilities that accompany this remarkable journey. Join us in unraveling the intricacies of AI’s evolution—a narrative that prompts reflection on the transformative power and potential of this groundbreaking technology. Don’t miss the chance to be part of this exploration into the unknown and prepare to be captivated by the unfolding story of AI’s adolescence.

  • | | |

    The Illusion of Intelligence: Why AI Hallucinations Still Exist

    AI hallucinations occur when language models generate convincing but false information. This happens because large language models are designed to predict plausible sequences of words rather than verify factual truth. Although new techniques are reducing errors, hallucinations remain a fundamental challenge of generative AI. As these systems become more persuasive, human judgment and critical thinking remain essential.

  • | | |

    The Amara Law

    The Amara law, which states that we tend to overestimate the impact of technology in the short term and underestimate it in the long term, is a well-known principle in the field of technology. As artificial intelligence (AI) continues to advance, it is important to consider how the Amara law applies to AI. In this paper, we explore the potential implications of the Amara law on the development and deployment of AI technologies, and discuss how stakeholders can leverage this principle to make better decisions about the use of AI.