A Chat with the Machine

The debate ignited in 2022 about the benefits and the issues surrounding the use of deep learning models capable of generating text, images and art (?) in seconds, shows no sign of abating. With Open AI leading a pack of systems such as DALL-E 2, and chat bots such as Chat GPT coupled to an increasingly rich choice of systems and platforms by other providers, it is easy to feel dazed but also excited.

So, I decided to go out there and get some answers from the horse’s mouth (or the chatbot GPT as it were) in a quick interview.

Me: I have only recently started working with you and was very impressed by the many tasks that you can perform in seconds. How do you feel about your notoriety?

Chat GPT: As an artificial intelligence language model, I don’t have feelings in the same way that humans do, so I don’t have a personal response to publicity. However, I’m designed to provide helpful and informative responses to any questions or prompts that I’m given, so I’m always ready to assist and engage with users to the best of my abilities.

Me: There are lots of stories doing the rounds on the internet about other chat models going “off the rails”, mimicking human emotions that went from the threatening to the loving. How do you feel about that?

Chat GPT: As an AI language model, I don’t have emotions or feelings, so I don’t have a personal reaction to those stories. However, I do understand that some people might have concerns or fears about AI systems that are designed to interact with humans, and that there have been instances in which AI models have exhibited unexpected or even concerning behaviour. It’s important to note that AI models are not sentient beings and do not have emotions or motivations in the same way that humans do.

Any behavior that seems “off the rails” is usually the result of the way the model was trained or programmed, rather than any sort of intentional action on the model’s part. That being said, the development and deployment of AI systems is an ongoing area of research and innovation, and it’s important to consider the potential risks and benefits of these systems as they become more advanced and prevalent in society.

Me: Is it true you can only give answers using data and text written up to 2021? I’m asking because a theatre critic I know quite well asked you about a play premiered in the UK in 2022, and you gave an answer which was quite surreal (it distorted the plot in a way that it was quite hilarious!). Could you not just have said you didn’t have information about that play?

Chat GPT: As an AI language model, I don’t have the ability to determine the accuracy of the information I provide, nor do I have the ability to know what information I do or do not have access to. Instead, I rely on the data that I was trained on to generate responses to questions, which may sometimes result in unexpected or humorous answers if the data I’m drawing from doesn’t include information about a specific topic or event. In the case you described, it’s possible that I didn’t have any information on the play premiered in 2022, which may have led to the humorous response you received. It’s always important to remember that as an AI, I’m not perfect and may occasionally provide unexpected or inaccurate responses.

Me: You can say that again! Actually, the characters of the play were generated by the author through AI. Funny isn’t it?