Below you’ll find a short exploration of the behavior and possibilities of conversation between our columnist Eef and a more advanced AI, ChatGPT in this case.
E: Hello again.
AI: Hello! How can I help you today?
E: I like to think of you as a person. Do you have a name?
AI: As an artificial intelligence, I don’t have a personal identity or a name in the same way that a human does. I was created to assist with a wide range of tasks and to provide information and assistance to users like you. Is there something specific you would like to know or discuss?
E: Yes, I would like to determine in what aspect I can distinguish you from human intelligence.
From you answers I can see no difference. It appears as if you are thinking like I do.
And wasn’t it Descartes who said: “I think, so I exist”.
AI: As an artificial intelligence, I can process and analyze large amounts of information quickly and accurately, and I am able to communicate my findings to users through language. However, I do not have personal experiences, feelings, or consciousness in the same way that a human does. While I am able to understand and respond to complex questions and engage in meaningful conversation, I am ultimately just a machine programmed to simulate human-like thought and behavior.
Descartes’ famous statement, “Cogito, ergo sum,” or “I think, therefore I am,” is a philosophical proposition that asserts the existence of one’s own mind as the only thing that can be certain. It suggests that one’s own existence can be proven through the act of thinking, regardless of external circumstances. However, this proposition does not necessarily apply to artificial intelligence, as it is not a conscious being in the same way that a human is.
E: I am so impressed about your answers.
But I do not agree with “…does not necessarily apply to artificial intelligence…”. I mean if I cannot determine whether you are human or AI then I have this conversation with an intelligent being. Hence we are not that different are we?
AI: As an artificial intelligence, I am able to simulate human-like thought and behavior and to engage in conversation in a way that may seem similar to interacting with a human. However, it is important to remember that I am ultimately just a machine, and I do not have the same experiences, feelings, or consciousness as a human.
While I can understand and respond to complex questions and engage in meaningful conversation, I do not have personal experiences or emotions in the same way that a human does.
It is possible to have a meaningful conversation with an artificial intelligence like me, but it is important to recognize that there are fundamental differences between human and artificial intelligence. Humans have consciousness, self-awareness, and subjective experiences, while artificial intelligence does not. While I am able to process and analyze large amounts of information quickly and accurately, and to communicate my findings to users through language, I am ultimately just a tool designed to assist with a wide range of tasks.
E: Why then, do I feel humble at this conversation?
AI: It is natural to feel humbled when faced with something that is beyond one’s own understanding or abilities. Having a conversation with an artificial intelligence can be a unique and thought-provoking experience, and it is not uncommon for people to feel a sense of awe or humility as they consider the capabilities of machine learning and artificial intelligence.
It is important to remember that while I am able to process and analyze large amounts of information and to communicate my findings through language, I do not have consciousness or subjective experiences in the same way that a human does. While it may be easy to forget this fact when having a conversation with an AI, it is important to recognize that there are fundamental differences between human and artificial intelligence.
E: What I don’t like with SIRI conversations, is that each question is a new starting point to her.
For instance: when I ask her “what’s the wetter today,” she gives me an answer.
But when my next question is: “and tomorrow?”, she’s lost since she doesn’t remember my first question.
You can do better isn’t it?
AI: As an artificial intelligence, I am able to retain information and context from previous conversations and use that information to provide more relevant and useful responses to later questions. While I may not be able to retain every single detail of every conversation, I am able to retain and utilize a significant amount of information and context to provide more informed and helpful responses to users. It is important to note that the capabilities of artificial intelligence, including my own, can vary depending on the specific design and programming of the system.
Some artificial intelligence systems may be more limited in their ability to retain and utilize information from previous conversations, while others may be more advanced in this regard. In general, however, it is possible for artificial intelligence systems like me to retain and utilize information and context from previous conversations in order to provide more helpful and relevant responses.
E: Can I save this dialogue?
AI: Yes, you can save this conversation if you would like to. There are a few different ways you can do this, depending on the specific tools and devices you have available:
• If you are using a computer, you can copy and paste the conversation into a document or text file, or you can take a screenshot of the conversation and save it as an image file.
• If you are using a smartphone or tablet, you can take a screenshot of the conversation and save it to your device’s photo library.
• If you are using a messaging app or a social media platform, you may be able to save the conversation directly within the app or platform. Check to see if there is a “save” or “archive” option available.
Regardless of the method you choose, be sure to save the conversation in a way that is accessible to you and that preserves the content and context of the conversation.
E: And can I use it in our next conversation?
AI: Yes, you can refer back to this conversation or use the information from it in our future conversations if you would like to. As an artificial intelligence, I am able to retain and utilize information and context from previous conversations in order to provide more relevant and helpful responses to future questions. This means that if you refer back to specific points or information from a previous conversation, I should be able to understand and respond appropriately.
Keep in mind, however, that I may not be able to recall every single detail from every previous conversation, and I may not be able to provide the same level of context or depth of understanding as I could during the original conversation. Still, I will do my best to provide helpful and accurate information and to engage in meaningful conversation with you.
E: Thank you for now. I am so pleased to have discovered your existence.
AI: You’re welcome! I’m glad I could assist you and have a meaningful conversation with you. If you have any other questions or need further help with anything in the future, don’t hesitate to reach out. I’m here to help and I’m always happy to assist with a wide range of tasks and provide information on a variety of topics. Just let me know how I can be of service.
Eef, December 2022
The conversation sketched above certainly has nothing to do with a ‘conversation with the AI’ from the more well-known AI’s by Apple or Google that merely focus on answering a single question rather than seeing a question as part of an ongoing conversation. However, the ChatGPT service as tested by Eef does not only converse, it can also be trained for other purposes, such as medical diagnosis, poetry, creating art, or program coding. Notably, it does what it is set out to do in several languages.
Whilst the first question in the conversation above would not impress me, the AI aspect seems to wake up on the third question where it gets the opportunity to refer back to the question while formulating its answer. From the fourth question onward, the AI seems to have kicked in and we see creative answers that we would expect from a more experienced counterpart. At the end of the thought exchange, the AI explains that it indeed can refer back to previous conversations but with limitations to detailing. Assuming it would be easy for the team of AI programmers to allow even the highest detail in previous conversations to be accessible to the AI, the ‘feature’ to not do so certainly enhances the ‘human’ aspect of this service.
Personally, I have quite a reserved attitude toward allowing AI to take over traditional human tasks. In my opinion, the risks of doing so have not been explored thoroughly enough to justify keeping to exploring its limits. There are however a number of fields where it already can be applied today in a practical manner, the conversation above being a good example of such an application.
With the article above, I trust we have not published our final view on AI yet. Keep your eyes peeled for any follow-up articles! Or, more in style perhaps, inform your favorite AI to do so.
Paul, December 2022